Skip to content

perf: sort in memory without concat but instead use modified arrow sort [WIP] - #25108

Closed
rluvaton wants to merge 3 commits into
apache:mainfrom
rluvaton:check-if-sorting-without-concat-is-better
Closed

perf: sort in memory without concat but instead use modified arrow sort [WIP]#25108
rluvaton wants to merge 3 commits into
apache:mainfrom
rluvaton:check-if-sorting-without-concat-is-better

Conversation

@rluvaton

@rluvaton rluvaton commented Sep 9, 2026 •
edited
Loading

Copy link
Copy Markdown
Member

I do not plan to merge it like this, since it duplicating a lot of code from arrow, but I'm doing it as a POC

Which issue does this PR close?

N/A

Rationale for this change

I wanna see if there is a perf benefit of avoiding concat and sort/sort_to_indecies + take and replacing it with modified sort to indices + interleave

What changes are included in this PR?

copied arrow sort and modified it to work on slice of input array and make it return Vec<(array_index, index_in_array)> for lexort_to_indices or return Vec for sort

What is the testing strategy for this PR?

Unit tests and existing tests

Are there any user-facing changes?

No

I want to run benchmarks the same benchmarks as:

to compare

@rluvaton

rluvaton commented Sep 9, 2026

Copy link
Copy Markdown
Member Author

run benchmarks

@adriangbot adriangbot mentioned this pull request Sep 9, 2026
Draft
@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5601497362-2250-7csl6 6.12.94+ #1 SMP Fri Jul 17 09:42:57 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing check-if-sorting-without-concat-is-better (bbc14e7) to a5c809f (merge-base) diff

Run configuration 
run benchmark clickbench_partitioned

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5601497362-2251-bn5hj 6.12.94+ #1 SMP Fri Jul 17 09:42:57 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing check-if-sorting-without-concat-is-better (bbc14e7) to a5c809f (merge-base) diff

Run configuration 
run benchmark tpcds

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5601497362-2252-4kzbw 6.12.94+ #1 SMP Fri Jul 17 09:42:57 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing check-if-sorting-without-concat-is-better (bbc14e7) to a5c809f (merge-base) diff

Run configuration 
run benchmark tpch

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

Benchmark for this request failed before finishing (Kubernetes reason: BackoffLimitExceeded).

Benchmarks requested: clickbench_partitioned

Runner log (last 40 lines) 
    |              |
    |              required by a bound introduced by this call
    |
help: the trait `FromIterator<Result<_, arrow_schema::ArrowError>>` is not implemented for `Result<Vec<_>, _>`
      but trait `FromIterator<Result<_, DataFusionError>>` is implemented for `Result<Vec<Arc<dyn arrow::array::Array>>, _>`
   --> /rustc/88d9e12ae178fab0fb5cc050a94da85685d449ea/library/core/src/result.rs:2114:0
    = help: for that trait implementation, expected `DataFusionError`, found `arrow_schema::ArrowError`
note: the method call chain might not have had the expected associated types
   --> datafusion/physical-plan/src/sorts/stream.rs:440:14
    |
439 |           let columns = (0..schema.fields().len())
    |                         -------------------------- this expression has type `Range<usize>`
440 |               .map(|i| {
    |  ______________^
441 | |                 let column_values: Vec<&dyn Array> = input_batches
442 | |                     .iter()
443 | |                     .map(|batch| batch.column(i).as_ref())
444 | |                     .collect();
445 | |                 interleave(&column_values, indices)
446 | |             })
    | |______________^ `Iterator::Item` is `Result<Arc<dyn Array>, ArrowError>` here
note: required by a bound in `std::iter::Iterator::collect`
   --> /rustc/88d9e12ae178fab0fb5cc050a94da85685d449ea/library/core/src/iter/traits/iterator.rs:2077:4

warning: variable does not need to be mutable
   --> datafusion/physical-plan/src/sorts/stream.rs:513:17
    |
513 |                 mut indices,
    |                 ----^^^^^^^
    |                 |
    |                 help: remove this `mut`
    |
    = note: `#[warn(unused_mut)]` (part of `#[warn(unused)]`) on by default

Some errors have detailed explanations: E0277, E0599, E0631.
For more information about an error, try `rustc --explain E0277`.
warning: `datafusion-physical-plan` (lib) generated 2 warnings
error: could not compile `datafusion-physical-plan` (lib) due to 4 previous errors; 2 warnings emitted
warning: build failed, waiting for other jobs to finish...
Kubernetes message 
Job has reached the specified backoff limit

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

Benchmark for this request failed before finishing (Kubernetes reason: BackoffLimitExceeded).

Benchmarks requested: tpcds

Runner log (last 40 lines) 
    |              |
    |              required by a bound introduced by this call
    |
help: the trait `FromIterator<Result<_, arrow_schema::ArrowError>>` is not implemented for `Result<Vec<_>, _>`
      but trait `FromIterator<Result<_, DataFusionError>>` is implemented for `Result<Vec<Arc<dyn arrow::array::Array>>, _>`
   --> /rustc/88d9e12ae178fab0fb5cc050a94da85685d449ea/library/core/src/result.rs:2114:0
    = help: for that trait implementation, expected `DataFusionError`, found `arrow_schema::ArrowError`
note: the method call chain might not have had the expected associated types
   --> datafusion/physical-plan/src/sorts/stream.rs:440:14
    |
439 |           let columns = (0..schema.fields().len())
    |                         -------------------------- this expression has type `Range<usize>`
440 |               .map(|i| {
    |  ______________^
441 | |                 let column_values: Vec<&dyn Array> = input_batches
442 | |                     .iter()
443 | |                     .map(|batch| batch.column(i).as_ref())
444 | |                     .collect();
445 | |                 interleave(&column_values, indices)
446 | |             })
    | |______________^ `Iterator::Item` is `Result<Arc<dyn Array>, ArrowError>` here
note: required by a bound in `std::iter::Iterator::collect`
   --> /rustc/88d9e12ae178fab0fb5cc050a94da85685d449ea/library/core/src/iter/traits/iterator.rs:2077:4

warning: variable does not need to be mutable
   --> datafusion/physical-plan/src/sorts/stream.rs:513:17
    |
513 |                 mut indices,
    |                 ----^^^^^^^
    |                 |
    |                 help: remove this `mut`
    |
    = note: `#[warn(unused_mut)]` (part of `#[warn(unused)]`) on by default

Some errors have detailed explanations: E0277, E0599, E0631.
For more information about an error, try `rustc --explain E0277`.
warning: `datafusion-physical-plan` (lib) generated 2 warnings
error: could not compile `datafusion-physical-plan` (lib) due to 4 previous errors; 2 warnings emitted
warning: build failed, waiting for other jobs to finish...
Kubernetes message 
Job has reached the specified backoff limit

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

Benchmark for this request failed before finishing (Kubernetes reason: BackoffLimitExceeded).

Benchmarks requested: tpch

Runner log (last 40 lines) 
    |              |
    |              required by a bound introduced by this call
    |
help: the trait `FromIterator<Result<_, arrow_schema::ArrowError>>` is not implemented for `Result<Vec<_>, _>`
      but trait `FromIterator<Result<_, DataFusionError>>` is implemented for `Result<Vec<Arc<dyn arrow::array::Array>>, _>`
   --> /rustc/88d9e12ae178fab0fb5cc050a94da85685d449ea/library/core/src/result.rs:2114:0
    = help: for that trait implementation, expected `DataFusionError`, found `arrow_schema::ArrowError`
note: the method call chain might not have had the expected associated types
   --> datafusion/physical-plan/src/sorts/stream.rs:440:14
    |
439 |           let columns = (0..schema.fields().len())
    |                         -------------------------- this expression has type `Range<usize>`
440 |               .map(|i| {
    |  ______________^
441 | |                 let column_values: Vec<&dyn Array> = input_batches
442 | |                     .iter()
443 | |                     .map(|batch| batch.column(i).as_ref())
444 | |                     .collect();
445 | |                 interleave(&column_values, indices)
446 | |             })
    | |______________^ `Iterator::Item` is `Result<Arc<dyn Array>, ArrowError>` here
note: required by a bound in `std::iter::Iterator::collect`
   --> /rustc/88d9e12ae178fab0fb5cc050a94da85685d449ea/library/core/src/iter/traits/iterator.rs:2077:4

warning: variable does not need to be mutable
   --> datafusion/physical-plan/src/sorts/stream.rs:513:17
    |
513 |                 mut indices,
    |                 ----^^^^^^^
    |                 |
    |                 help: remove this `mut`
    |
    = note: `#[warn(unused_mut)]` (part of `#[warn(unused)]`) on by default

Some errors have detailed explanations: E0277, E0599, E0631.
For more information about an error, try `rustc --explain E0277`.
warning: `datafusion-physical-plan` (lib) generated 2 warnings
error: could not compile `datafusion-physical-plan` (lib) due to 4 previous errors; 2 warnings emitted
warning: build failed, waiting for other jobs to finish...
Kubernetes message 
Job has reached the specified backoff limit

File an issue against this benchmark runner

@rluvaton

rluvaton commented Sep 9, 2026

Copy link
Copy Markdown
Member Author

run benchmarks

@github-actions github-actions Bot added the physical-plan Changes to the physical-plan crate label Sep 9, 2026
@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5601774975-2253-95zp8 6.12.94+ #1 SMP Fri Jul 17 09:42:57 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark clickbench_partitioned

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5601774975-2254-6kpc8 6.12.94+ #1 SMP Fri Jul 17 09:42:57 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark tpcds

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5601774975-2255-7clq6 6.12.94+ #1 SMP Fri Jul 17 09:42:57 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark tpch

Results will be posted here when complete

File an issue against this benchmark runner

@rluvaton

rluvaton commented Sep 9, 2026

Copy link
Copy Markdown
Member Author

run benchmarks tpch tpch10 tpcds

env:
   prefer_hash_join: false

@rluvaton

rluvaton commented Sep 9, 2026

Copy link
Copy Markdown
Member Author

run benchmark sort_tpch10

@rluvaton

rluvaton commented Sep 9, 2026

Copy link
Copy Markdown
Member Author

run benchmark sort_tpch

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5601904691-2256-5dwnk 6.12.94+ #1 SMP Fri Jul 17 09:42:57 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark tpch
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5601904691-2257-jbh79 6.12.94+ #1 SMP Fri Jul 17 09:42:57 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark tpch10
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5601904691-2258-4k658 6.12.94+ #1 SMP Fri Jul 17 09:42:57 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark tpcds
env:
  prefer_hash_join: "false"

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark tpch
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃     HEAD ┃ check-if-sorting-without-concat-is-better ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 39.44 ms │                                  38.24 ms │ no change │
│ QQuery 2  │ 19.48 ms │                                  19.21 ms │ no change │
│ QQuery 3  │ 29.62 ms │                                  29.34 ms │ no change │
│ QQuery 4  │ 18.21 ms │                                  18.25 ms │ no change │
│ QQuery 5  │ 37.50 ms │                                  36.92 ms │ no change │
│ QQuery 6  │ 16.54 ms │                                  16.41 ms │ no change │
│ QQuery 7  │ 42.73 ms │                                  42.82 ms │ no change │
│ QQuery 8  │ 42.15 ms │                                  41.91 ms │ no change │
│ QQuery 9  │ 51.60 ms │                                  50.29 ms │ no change │
│ QQuery 10 │ 43.64 ms │                                  43.25 ms │ no change │
│ QQuery 11 │ 14.40 ms │                                  14.23 ms │ no change │
│ QQuery 12 │ 24.45 ms │                                  24.67 ms │ no change │
│ QQuery 13 │ 41.86 ms │                                  41.96 ms │ no change │
│ QQuery 14 │ 25.24 ms │                                  25.07 ms │ no change │
│ QQuery 15 │ 31.68 ms │                                  31.55 ms │ no change │
│ QQuery 16 │ 14.40 ms │                                  14.15 ms │ no change │
│ QQuery 17 │ 73.50 ms │                                  73.55 ms │ no change │
│ QQuery 18 │ 61.34 ms │                                  62.01 ms │ no change │
│ QQuery 19 │ 33.23 ms │                                  33.23 ms │ no change │
│ QQuery 20 │ 31.66 ms │                                  32.21 ms │ no change │
│ QQuery 21 │ 57.37 ms │                                  57.60 ms │ no change │
│ QQuery 22 │ 14.50 ms │                                  14.93 ms │ no change │
└───────────┴──────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 764.54ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 761.80ms │
│ Average Time (HEAD)                                      │  34.75ms │
│ Average Time (check-if-sorting-without-concat-is-better) │  34.63ms │
│ Queries Faster                                           │        0 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       22 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃ check-if-sorting-without-concat-is-better ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 39.44 / 41.06 ±1.72 / 43.75 ms │            38.24 / 39.54 ±1.19 / 41.30 ms │     no change │
│ QQuery 2  │ 19.48 / 20.36 ±0.91 / 21.56 ms │            19.21 / 19.66 ±0.49 / 20.60 ms │     no change │
│ QQuery 3  │ 29.62 / 30.41 ±1.12 / 32.58 ms │            29.34 / 30.02 ±0.54 / 30.87 ms │     no change │
│ QQuery 4  │ 18.21 / 20.08 ±1.16 / 21.84 ms │            18.25 / 18.83 ±0.53 / 19.67 ms │ +1.07x faster │
│ QQuery 5  │ 37.50 / 37.97 ±0.34 / 38.43 ms │            36.92 / 38.02 ±1.11 / 39.94 ms │     no change │
│ QQuery 6  │ 16.54 / 17.45 ±1.10 / 18.98 ms │            16.41 / 16.46 ±0.06 / 16.57 ms │ +1.06x faster │
│ QQuery 7  │ 42.73 / 43.99 ±0.82 / 45.00 ms │            42.82 / 45.04 ±2.95 / 50.86 ms │     no change │
│ QQuery 8  │ 42.15 / 42.88 ±0.75 / 44.31 ms │            41.91 / 42.47 ±0.55 / 43.45 ms │     no change │
│ QQuery 9  │ 51.60 / 52.62 ±0.94 / 54.38 ms │            50.29 / 52.17 ±1.29 / 53.69 ms │     no change │
│ QQuery 10 │ 43.64 / 44.22 ±0.51 / 45.07 ms │            43.25 / 43.80 ±0.76 / 45.32 ms │     no change │
│ QQuery 11 │ 14.40 / 14.75 ±0.54 / 15.82 ms │            14.23 / 14.63 ±0.51 / 15.64 ms │     no change │
│ QQuery 12 │ 24.45 / 25.12 ±0.41 / 25.60 ms │            24.67 / 25.00 ±0.38 / 25.74 ms │     no change │
│ QQuery 13 │ 41.86 / 43.76 ±2.02 / 47.13 ms │            41.96 / 42.84 ±1.06 / 44.79 ms │     no change │
│ QQuery 14 │ 25.24 / 25.41 ±0.16 / 25.64 ms │            25.07 / 25.32 ±0.28 / 25.68 ms │     no change │
│ QQuery 15 │ 31.68 / 32.38 ±0.69 / 33.63 ms │            31.55 / 32.07 ±0.40 / 32.54 ms │     no change │
│ QQuery 16 │ 14.40 / 14.64 ±0.20 / 14.98 ms │            14.15 / 14.40 ±0.25 / 14.83 ms │     no change │
│ QQuery 17 │ 73.50 / 74.86 ±0.85 / 76.02 ms │            73.55 / 74.62 ±0.95 / 76.31 ms │     no change │
│ QQuery 18 │ 61.34 / 62.11 ±0.58 / 62.91 ms │            62.01 / 62.71 ±0.82 / 64.06 ms │     no change │
│ QQuery 19 │ 33.23 / 33.43 ±0.25 / 33.91 ms │            33.23 / 33.80 ±0.47 / 34.53 ms │     no change │
│ QQuery 20 │ 31.66 / 32.07 ±0.30 / 32.52 ms │            32.21 / 32.55 ±0.18 / 32.71 ms │     no change │
│ QQuery 21 │ 57.37 / 58.47 ±1.44 / 61.32 ms │            57.60 / 58.22 ±0.48 / 59.01 ms │     no change │
│ QQuery 22 │ 14.50 / 15.17 ±0.94 / 17.00 ms │            14.93 / 15.02 ±0.09 / 15.16 ms │     no change │
└───────────┴────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 783.21ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 777.21ms │
│ Average Time (HEAD)                                      │  35.60ms │
│ Average Time (check-if-sorting-without-concat-is-better) │  35.33ms │
│ Queries Faster                                           │        2 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       20 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 1.2 GiB
Avg memory 492.7 MiB
CPU user 22.1s
CPU sys 1.8s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.2 GiB
Avg memory 513.2 MiB
CPU user 22.0s
CPU sys 1.8s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark tpcds
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ check-if-sorting-without-concat-is-better ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.87 ms │                                   5.70 ms │     no change │
│ QQuery 2  │   80.34 ms │                                  80.62 ms │     no change │
│ QQuery 3  │   29.29 ms │                                  29.02 ms │     no change │
│ QQuery 4  │  486.11 ms │                                 483.46 ms │     no change │
│ QQuery 5  │   52.21 ms │                                  52.01 ms │     no change │
│ QQuery 6  │   36.42 ms │                                  36.28 ms │     no change │
│ QQuery 7  │   75.05 ms │                                  75.11 ms │     no change │
│ QQuery 8  │   36.43 ms │                                  36.38 ms │     no change │
│ QQuery 9  │   52.28 ms │                                  52.01 ms │     no change │
│ QQuery 10 │   63.68 ms │                                  62.67 ms │     no change │
│ QQuery 11 │  299.11 ms │                                 302.47 ms │     no change │
│ QQuery 12 │   28.77 ms │                                  28.53 ms │     no change │
│ QQuery 13 │  118.13 ms │                                 117.82 ms │     no change │
│ QQuery 14 │  415.92 ms │                                 415.94 ms │     no change │
│ QQuery 15 │   57.38 ms │                                  57.00 ms │     no change │
│ QQuery 16 │    6.95 ms │                                   6.99 ms │     no change │
│ QQuery 17 │   80.23 ms │                                  80.23 ms │     no change │
│ QQuery 18 │  104.98 ms │                                 104.30 ms │     no change │
│ QQuery 19 │   41.09 ms │                                  41.53 ms │     no change │
│ QQuery 20 │   35.78 ms │                                  35.19 ms │     no change │
│ QQuery 21 │   17.49 ms │                                  17.41 ms │     no change │
│ QQuery 22 │   63.40 ms │                                  62.84 ms │     no change │
│ QQuery 23 │  309.03 ms │                                 311.30 ms │     no change │
│ QQuery 24 │  195.95 ms │                                 193.41 ms │     no change │
│ QQuery 25 │  109.02 ms │                                 109.21 ms │     no change │
│ QQuery 26 │   49.34 ms │                                  48.47 ms │     no change │
│ QQuery 27 │    6.35 ms │                                   6.08 ms │     no change │
│ QQuery 28 │   61.00 ms │                                  55.95 ms │ +1.09x faster │
│ QQuery 29 │   97.42 ms │                                  96.79 ms │     no change │
│ QQuery 30 │   32.11 ms │                                  32.00 ms │     no change │
│ QQuery 31 │  110.24 ms │                                 110.95 ms │     no change │
│ QQuery 32 │   20.76 ms │                                  20.11 ms │     no change │
│ QQuery 33 │   37.95 ms │                                  38.13 ms │     no change │
│ QQuery 34 │    9.87 ms │                                   9.97 ms │     no change │
│ QQuery 35 │   72.42 ms │                                  72.54 ms │     no change │
│ QQuery 36 │    5.99 ms │                                   5.87 ms │     no change │
│ QQuery 37 │    6.97 ms │                                   6.74 ms │     no change │
│ QQuery 38 │   61.60 ms │                                  62.48 ms │     no change │
│ QQuery 39 │   88.98 ms │                                  89.20 ms │     no change │
│ QQuery 40 │   23.99 ms │                                  24.27 ms │     no change │
│ QQuery 41 │   11.12 ms │                                  11.13 ms │     no change │
│ QQuery 42 │   23.84 ms │                                  23.56 ms │     no change │
│ QQuery 43 │    5.37 ms │                                   5.42 ms │     no change │
│ QQuery 44 │    9.49 ms │                                   9.49 ms │     no change │
│ QQuery 45 │   38.84 ms │                                  38.40 ms │     no change │
│ QQuery 46 │   12.06 ms │                                  12.17 ms │     no change │
│ QQuery 47 │  225.50 ms │                                 224.66 ms │     no change │
│ QQuery 48 │   95.08 ms │                                  95.36 ms │     no change │
│ QQuery 49 │   71.59 ms │                                  72.10 ms │     no change │
│ QQuery 50 │   58.79 ms │                                  59.46 ms │     no change │
│ QQuery 51 │   92.18 ms │                                  91.45 ms │     no change │
│ QQuery 52 │   23.95 ms │                                  24.08 ms │     no change │
│ QQuery 53 │   29.00 ms │                                  28.94 ms │     no change │
│ QQuery 54 │   54.09 ms │                                  54.62 ms │     no change │
│ QQuery 55 │   23.47 ms │                                  23.44 ms │     no change │
│ QQuery 56 │   39.38 ms │                                  39.20 ms │     no change │
│ QQuery 57 │  173.73 ms │                                 175.35 ms │     no change │
│ QQuery 58 │  111.81 ms │                                 111.40 ms │     no change │
│ QQuery 59 │  117.07 ms │                                 117.61 ms │     no change │
│ QQuery 60 │   39.51 ms │                                  39.25 ms │     no change │
│ QQuery 61 │   12.42 ms │                                  12.30 ms │     no change │
│ QQuery 62 │   46.24 ms │                                  45.87 ms │     no change │
│ QQuery 63 │   29.29 ms │                                  29.09 ms │     no change │
│ QQuery 64 │  366.83 ms │                                 364.55 ms │     no change │
│ QQuery 65 │  121.77 ms │                                 120.20 ms │     no change │
│ QQuery 66 │   82.17 ms │                                  80.89 ms │     no change │
│ QQuery 67 │  241.44 ms │                                 242.55 ms │     no change │
│ QQuery 68 │   12.02 ms │                                  11.99 ms │     no change │
│ QQuery 69 │   58.26 ms │                                  57.01 ms │     no change │
│ QQuery 70 │  105.28 ms │                                 104.88 ms │     no change │
│ QQuery 71 │   35.72 ms │                                  35.32 ms │     no change │
│ QQuery 72 │ 1766.09 ms │                                1889.38 ms │  1.07x slower │
│ QQuery 73 │    9.91 ms │                                  10.04 ms │     no change │
│ QQuery 74 │  170.51 ms │                                 170.01 ms │     no change │
│ QQuery 75 │  146.27 ms │                                 145.43 ms │     no change │
│ QQuery 76 │   35.07 ms │                                  35.54 ms │     no change │
│ QQuery 77 │   61.70 ms │                                  60.99 ms │     no change │
│ QQuery 78 │  219.48 ms │                                 220.50 ms │     no change │
│ QQuery 79 │   65.96 ms │                                  66.59 ms │     no change │
│ QQuery 80 │   98.46 ms │                                  99.66 ms │     no change │
│ QQuery 81 │   25.68 ms │                                  25.83 ms │     no change │
│ QQuery 82 │   16.20 ms │                                  16.22 ms │     no change │
│ QQuery 83 │   33.92 ms │                                  34.06 ms │     no change │
│ QQuery 84 │   29.56 ms │                                  29.52 ms │     no change │
│ QQuery 85 │  103.03 ms │                                 102.46 ms │     no change │
│ QQuery 86 │   25.07 ms │                                  25.21 ms │     no change │
│ QQuery 87 │   62.21 ms │                                  62.25 ms │     no change │
│ QQuery 88 │   62.80 ms │                                  63.28 ms │     no change │
│ QQuery 89 │   35.42 ms │                                  35.66 ms │     no change │
│ QQuery 90 │   16.99 ms │                                  17.05 ms │     no change │
│ QQuery 91 │   44.78 ms │                                  44.81 ms │     no change │
│ QQuery 92 │   29.13 ms │                                  28.91 ms │     no change │
│ QQuery 93 │   49.15 ms │                                  50.19 ms │     no change │
│ QQuery 94 │   37.78 ms │                                  37.85 ms │     no change │
│ QQuery 95 │   80.30 ms │                                  80.70 ms │     no change │
│ QQuery 96 │   23.76 ms │                                  23.70 ms │     no change │
│ QQuery 97 │   52.21 ms │                                  52.71 ms │     no change │
│ QQuery 98 │   43.07 ms │                                  42.27 ms │     no change │
│ QQuery 99 │   69.60 ms │                                  69.49 ms │     no change │
└───────────┴────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 9267.29ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 9379.08ms │
│ Average Time (HEAD)                                      │   93.61ms │
│ Average Time (check-if-sorting-without-concat-is-better) │   94.74ms │
│ Queries Faster                                           │         1 │
│ Queries Slower                                           │         1 │
│ Queries with No Change                                   │        97 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ check-if-sorting-without-concat-is-better ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.87 / 6.44 ±0.99 / 8.41 ms │               5.70 / 6.23 ±0.95 / 8.13 ms │     no change │
│ QQuery 2  │        80.34 / 80.73 ±0.29 / 81.17 ms │            80.62 / 81.03 ±0.21 / 81.21 ms │     no change │
│ QQuery 3  │        29.29 / 29.42 ±0.08 / 29.50 ms │            29.02 / 29.19 ±0.11 / 29.34 ms │     no change │
│ QQuery 4  │     486.11 / 488.90 ±1.48 / 490.47 ms │         483.46 / 487.95 ±3.16 / 491.79 ms │     no change │
│ QQuery 5  │        52.21 / 52.73 ±0.30 / 53.05 ms │            52.01 / 52.89 ±0.53 / 53.43 ms │     no change │
│ QQuery 6  │        36.42 / 36.80 ±0.33 / 37.22 ms │            36.28 / 36.69 ±0.34 / 37.08 ms │     no change │
│ QQuery 7  │        75.05 / 75.55 ±0.30 / 75.94 ms │            75.11 / 75.67 ±0.33 / 76.12 ms │     no change │
│ QQuery 8  │        36.43 / 38.08 ±2.23 / 42.48 ms │            36.38 / 36.80 ±0.23 / 37.03 ms │     no change │
│ QQuery 9  │        52.28 / 54.35 ±2.73 / 59.69 ms │            52.01 / 53.53 ±1.52 / 56.43 ms │     no change │
│ QQuery 10 │        63.68 / 63.92 ±0.22 / 64.24 ms │            62.67 / 63.01 ±0.32 / 63.61 ms │     no change │
│ QQuery 11 │     299.11 / 302.65 ±2.73 / 306.93 ms │         302.47 / 303.93 ±1.19 / 305.98 ms │     no change │
│ QQuery 12 │        28.77 / 29.18 ±0.38 / 29.71 ms │            28.53 / 28.94 ±0.29 / 29.36 ms │     no change │
│ QQuery 13 │     118.13 / 118.58 ±0.42 / 119.26 ms │         117.82 / 118.61 ±0.75 / 119.92 ms │     no change │
│ QQuery 14 │     415.92 / 418.35 ±1.61 / 420.22 ms │         415.94 / 420.25 ±3.56 / 425.06 ms │     no change │
│ QQuery 15 │        57.38 / 58.84 ±1.44 / 61.40 ms │            57.00 / 57.58 ±0.42 / 58.15 ms │     no change │
│ QQuery 16 │           6.95 / 7.07 ±0.21 / 7.49 ms │               6.99 / 7.20 ±0.24 / 7.67 ms │     no change │
│ QQuery 17 │        80.23 / 82.11 ±2.46 / 86.76 ms │            80.23 / 80.96 ±0.92 / 82.69 ms │     no change │
│ QQuery 18 │     104.98 / 106.51 ±1.35 / 108.64 ms │         104.30 / 105.10 ±0.69 / 106.22 ms │     no change │
│ QQuery 19 │        41.09 / 41.74 ±0.67 / 43.00 ms │            41.53 / 41.63 ±0.08 / 41.74 ms │     no change │
│ QQuery 20 │        35.78 / 36.04 ±0.17 / 36.30 ms │            35.19 / 36.12 ±0.74 / 36.99 ms │     no change │
│ QQuery 21 │        17.49 / 18.23 ±1.05 / 20.31 ms │            17.41 / 17.59 ±0.14 / 17.82 ms │     no change │
│ QQuery 22 │        63.40 / 64.01 ±0.55 / 64.90 ms │            62.84 / 64.00 ±0.65 / 64.79 ms │     no change │
│ QQuery 23 │     309.03 / 313.75 ±2.61 / 316.85 ms │         311.30 / 314.17 ±3.32 / 320.38 ms │     no change │
│ QQuery 24 │     195.95 / 200.35 ±4.86 / 209.40 ms │         193.41 / 196.87 ±2.53 / 200.37 ms │     no change │
│ QQuery 25 │     109.02 / 112.30 ±3.53 / 119.03 ms │         109.21 / 110.96 ±1.18 / 112.59 ms │     no change │
│ QQuery 26 │        49.34 / 49.59 ±0.23 / 49.88 ms │            48.47 / 48.83 ±0.33 / 49.33 ms │     no change │
│ QQuery 27 │           6.35 / 6.56 ±0.24 / 7.02 ms │               6.08 / 6.25 ±0.17 / 6.55 ms │     no change │
│ QQuery 28 │        61.00 / 62.22 ±1.39 / 64.94 ms │            55.95 / 59.13 ±2.32 / 61.72 ms │     no change │
│ QQuery 29 │       97.42 / 99.81 ±2.79 / 105.26 ms │            96.79 / 97.85 ±0.85 / 99.03 ms │     no change │
│ QQuery 30 │        32.11 / 32.49 ±0.30 / 32.82 ms │            32.00 / 32.55 ±0.36 / 32.95 ms │     no change │
│ QQuery 31 │     110.24 / 112.63 ±2.48 / 117.31 ms │         110.95 / 112.73 ±1.93 / 115.68 ms │     no change │
│ QQuery 32 │        20.76 / 21.00 ±0.35 / 21.70 ms │            20.11 / 20.57 ±0.28 / 20.97 ms │     no change │
│ QQuery 33 │        37.95 / 38.50 ±0.30 / 38.84 ms │            38.13 / 38.86 ±1.05 / 40.93 ms │     no change │
│ QQuery 34 │         9.87 / 10.15 ±0.21 / 10.40 ms │             9.97 / 10.38 ±0.40 / 11.07 ms │     no change │
│ QQuery 35 │        72.42 / 72.83 ±0.34 / 73.43 ms │            72.54 / 73.66 ±1.20 / 75.98 ms │     no change │
│ QQuery 36 │           5.99 / 6.13 ±0.15 / 6.42 ms │               5.87 / 5.99 ±0.21 / 6.41 ms │     no change │
│ QQuery 37 │          6.97 / 7.79 ±1.27 / 10.31 ms │               6.74 / 6.90 ±0.14 / 7.09 ms │ +1.13x faster │
│ QQuery 38 │        61.60 / 62.48 ±0.72 / 63.51 ms │            62.48 / 62.98 ±0.55 / 64.02 ms │     no change │
│ QQuery 39 │        88.98 / 90.39 ±1.07 / 91.98 ms │            89.20 / 91.48 ±2.13 / 95.01 ms │     no change │
│ QQuery 40 │        23.99 / 24.34 ±0.22 / 24.66 ms │            24.27 / 25.19 ±1.28 / 27.73 ms │     no change │
│ QQuery 41 │        11.12 / 11.37 ±0.18 / 11.68 ms │            11.13 / 11.51 ±0.43 / 12.28 ms │     no change │
│ QQuery 42 │        23.84 / 24.64 ±0.50 / 25.24 ms │            23.56 / 24.17 ±0.61 / 25.11 ms │     no change │
│ QQuery 43 │           5.37 / 5.46 ±0.12 / 5.68 ms │               5.42 / 5.53 ±0.14 / 5.80 ms │     no change │
│ QQuery 44 │           9.49 / 9.57 ±0.07 / 9.67 ms │               9.49 / 9.62 ±0.12 / 9.80 ms │     no change │
│ QQuery 45 │        38.84 / 40.06 ±1.30 / 42.56 ms │            38.40 / 39.56 ±0.65 / 40.22 ms │     no change │
│ QQuery 46 │        12.06 / 12.45 ±0.42 / 13.13 ms │            12.17 / 12.53 ±0.30 / 12.95 ms │     no change │
│ QQuery 47 │     225.50 / 228.49 ±2.75 / 233.24 ms │         224.66 / 228.16 ±4.17 / 235.63 ms │     no change │
│ QQuery 48 │        95.08 / 95.96 ±0.64 / 96.65 ms │            95.36 / 96.19 ±0.56 / 97.11 ms │     no change │
│ QQuery 49 │        71.59 / 72.72 ±0.63 / 73.42 ms │            72.10 / 72.78 ±0.58 / 73.65 ms │     no change │
│ QQuery 50 │        58.79 / 59.18 ±0.38 / 59.67 ms │            59.46 / 59.74 ±0.25 / 60.20 ms │     no change │
│ QQuery 51 │       92.18 / 95.26 ±4.05 / 103.26 ms │           91.45 / 94.49 ±4.46 / 103.32 ms │     no change │
│ QQuery 52 │        23.95 / 24.16 ±0.18 / 24.45 ms │            24.08 / 24.31 ±0.27 / 24.80 ms │     no change │
│ QQuery 53 │        29.00 / 29.68 ±0.54 / 30.61 ms │            28.94 / 29.63 ±0.99 / 31.59 ms │     no change │
│ QQuery 54 │        54.09 / 54.68 ±0.34 / 55.13 ms │            54.62 / 55.01 ±0.21 / 55.23 ms │     no change │
│ QQuery 55 │        23.47 / 23.86 ±0.29 / 24.28 ms │            23.44 / 23.64 ±0.11 / 23.76 ms │     no change │
│ QQuery 56 │        39.38 / 40.97 ±2.76 / 46.48 ms │            39.20 / 40.70 ±2.32 / 45.31 ms │     no change │
│ QQuery 57 │     173.73 / 176.29 ±1.35 / 177.41 ms │         175.35 / 177.96 ±2.28 / 181.79 ms │     no change │
│ QQuery 58 │     111.81 / 112.70 ±0.77 / 114.02 ms │         111.40 / 113.64 ±1.78 / 116.45 ms │     no change │
│ QQuery 59 │     117.07 / 118.23 ±1.74 / 121.69 ms │         117.61 / 118.76 ±1.17 / 120.96 ms │     no change │
│ QQuery 60 │        39.51 / 40.01 ±0.41 / 40.53 ms │            39.25 / 39.98 ±0.65 / 40.86 ms │     no change │
│ QQuery 61 │        12.42 / 13.12 ±1.13 / 15.36 ms │            12.30 / 12.54 ±0.15 / 12.69 ms │     no change │
│ QQuery 62 │        46.24 / 46.57 ±0.43 / 47.36 ms │            45.87 / 46.84 ±1.06 / 48.85 ms │     no change │
│ QQuery 63 │        29.29 / 29.83 ±0.44 / 30.40 ms │            29.09 / 29.38 ±0.23 / 29.62 ms │     no change │
│ QQuery 64 │     366.83 / 369.08 ±1.23 / 370.55 ms │         364.55 / 366.92 ±3.13 / 372.98 ms │     no change │
│ QQuery 65 │     121.77 / 125.03 ±2.10 / 127.51 ms │         120.20 / 122.06 ±1.59 / 124.58 ms │     no change │
│ QQuery 66 │        82.17 / 83.82 ±1.37 / 86.15 ms │            80.89 / 82.02 ±0.71 / 83.02 ms │     no change │
│ QQuery 67 │     241.44 / 243.58 ±1.83 / 246.12 ms │         242.55 / 245.27 ±2.60 / 249.10 ms │     no change │
│ QQuery 68 │        12.02 / 12.20 ±0.15 / 12.47 ms │            11.99 / 12.15 ±0.13 / 12.37 ms │     no change │
│ QQuery 69 │        58.26 / 59.03 ±0.82 / 60.09 ms │            57.01 / 58.73 ±2.47 / 63.64 ms │     no change │
│ QQuery 70 │     105.28 / 107.90 ±1.70 / 110.38 ms │         104.88 / 106.72 ±2.37 / 111.27 ms │     no change │
│ QQuery 71 │        35.72 / 36.19 ±0.34 / 36.62 ms │            35.32 / 35.75 ±0.43 / 36.43 ms │     no change │
│ QQuery 72 │ 1766.09 / 1847.37 ±61.52 / 1954.28 ms │     1889.38 / 1915.97 ±18.93 / 1934.98 ms │     no change │
│ QQuery 73 │         9.91 / 10.16 ±0.29 / 10.57 ms │            10.04 / 10.34 ±0.20 / 10.64 ms │     no change │
│ QQuery 74 │     170.51 / 173.31 ±2.19 / 176.38 ms │         170.01 / 173.10 ±2.40 / 177.02 ms │     no change │
│ QQuery 75 │     146.27 / 148.26 ±1.41 / 150.29 ms │         145.43 / 147.86 ±1.49 / 150.07 ms │     no change │
│ QQuery 76 │        35.07 / 35.66 ±0.57 / 36.67 ms │            35.54 / 35.93 ±0.29 / 36.32 ms │     no change │
│ QQuery 77 │        61.70 / 61.99 ±0.20 / 62.24 ms │            60.99 / 61.64 ±0.58 / 62.72 ms │     no change │
│ QQuery 78 │     219.48 / 228.52 ±5.05 / 234.70 ms │         220.50 / 224.12 ±2.43 / 227.50 ms │     no change │
│ QQuery 79 │        65.96 / 67.66 ±2.40 / 72.37 ms │            66.59 / 68.65 ±3.71 / 76.05 ms │     no change │
│ QQuery 80 │      98.46 / 101.23 ±2.05 / 104.70 ms │          99.66 / 100.50 ±0.88 / 102.08 ms │     no change │
│ QQuery 81 │        25.68 / 25.95 ±0.19 / 26.21 ms │            25.83 / 26.02 ±0.21 / 26.38 ms │     no change │
│ QQuery 82 │        16.20 / 16.36 ±0.19 / 16.74 ms │            16.22 / 16.47 ±0.20 / 16.72 ms │     no change │
│ QQuery 83 │        33.92 / 34.26 ±0.33 / 34.89 ms │            34.06 / 34.36 ±0.24 / 34.68 ms │     no change │
│ QQuery 84 │        29.56 / 30.89 ±2.45 / 35.79 ms │            29.52 / 31.62 ±2.80 / 37.14 ms │     no change │
│ QQuery 85 │     103.03 / 104.19 ±1.55 / 107.19 ms │         102.46 / 103.60 ±1.51 / 106.57 ms │     no change │
│ QQuery 86 │        25.07 / 25.52 ±0.51 / 26.50 ms │            25.21 / 25.47 ±0.17 / 25.71 ms │     no change │
│ QQuery 87 │        62.21 / 63.63 ±1.49 / 66.30 ms │            62.25 / 63.53 ±2.40 / 68.33 ms │     no change │
│ QQuery 88 │        62.80 / 64.59 ±1.32 / 66.67 ms │            63.28 / 64.04 ±0.49 / 64.66 ms │     no change │
│ QQuery 89 │        35.42 / 35.95 ±0.56 / 36.89 ms │            35.66 / 36.41 ±0.68 / 37.51 ms │     no change │
│ QQuery 90 │        16.99 / 17.22 ±0.14 / 17.40 ms │            17.05 / 17.18 ±0.11 / 17.34 ms │     no change │
│ QQuery 91 │        44.78 / 45.89 ±1.01 / 47.78 ms │            44.81 / 45.18 ±0.36 / 45.74 ms │     no change │
│ QQuery 92 │        29.13 / 30.01 ±0.65 / 31.12 ms │            28.91 / 31.63 ±3.72 / 39.00 ms │  1.05x slower │
│ QQuery 93 │        49.15 / 50.32 ±0.82 / 51.47 ms │            50.19 / 51.18 ±0.68 / 52.08 ms │     no change │
│ QQuery 94 │        37.78 / 39.27 ±2.17 / 43.59 ms │            37.85 / 39.04 ±1.86 / 42.74 ms │     no change │
│ QQuery 95 │        80.30 / 81.61 ±0.94 / 83.19 ms │            80.70 / 81.92 ±0.71 / 82.66 ms │     no change │
│ QQuery 96 │        23.76 / 24.13 ±0.19 / 24.28 ms │            23.70 / 24.04 ±0.24 / 24.33 ms │     no change │
│ QQuery 97 │        52.21 / 53.42 ±1.15 / 55.33 ms │            52.71 / 54.00 ±1.44 / 56.52 ms │     no change │
│ QQuery 98 │        43.07 / 44.52 ±1.53 / 47.30 ms │            42.27 / 43.44 ±1.08 / 45.43 ms │     no change │
│ QQuery 99 │        69.60 / 70.39 ±1.01 / 72.32 ms │            69.49 / 70.15 ±0.51 / 71.02 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 9467.94ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 9513.92ms │
│ Average Time (HEAD)                                      │   95.64ms │
│ Average Time (check-if-sorting-without-concat-is-better) │   96.10ms │
│ Queries Faster                                           │         1 │
│ Queries Slower                                           │         1 │
│ Queries with No Change                                   │        97 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 50.0s
Peak memory 2.3 GiB
Avg memory 1.6 GiB
CPU user 202.3s
CPU sys 5.6s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 50.0s
Peak memory 2.1 GiB
Avg memory 1.4 GiB
CPU user 207.0s
CPU sys 5.3s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5601909824-2260-tbhdr 6.12.94+ #1 SMP Fri Jul 17 09:42:57 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark sort_tpch

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5601906066-2259-8kpkk 6.12.94+ #1 SMP Fri Jul 17 09:42:57 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark sort_tpch10

Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark clickbench_partitioned
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ check-if-sorting-without-concat-is-better ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │    1.32 ms │                                   1.39 ms │     no change │
│ QQuery 1  │   12.03 ms │                                  12.19 ms │     no change │
│ QQuery 2  │   37.29 ms │                                  37.21 ms │     no change │
│ QQuery 3  │   30.65 ms │                                  32.35 ms │  1.06x slower │
│ QQuery 4  │  222.84 ms │                                 265.69 ms │  1.19x slower │
│ QQuery 5  │  275.33 ms │                                 305.32 ms │  1.11x slower │
│ QQuery 6  │    1.28 ms │                                   1.46 ms │  1.14x slower │
│ QQuery 7  │   13.12 ms │                                  13.41 ms │     no change │
│ QQuery 8  │  327.79 ms │                                 326.34 ms │     no change │
│ QQuery 9  │  451.64 ms │                                 449.60 ms │     no change │
│ QQuery 10 │   70.83 ms │                                  68.94 ms │     no change │
│ QQuery 11 │   81.97 ms │                                  84.87 ms │     no change │
│ QQuery 12 │  269.07 ms │                                 296.11 ms │  1.10x slower │
│ QQuery 13 │  369.41 ms │                                 405.38 ms │  1.10x slower │
│ QQuery 14 │  284.39 ms │                                 283.92 ms │     no change │
│ QQuery 15 │  269.64 ms │                                 271.91 ms │     no change │
│ QQuery 16 │  607.64 ms │                                 615.34 ms │     no change │
│ QQuery 17 │  626.31 ms │                                 637.05 ms │     no change │
│ QQuery 18 │ 1263.61 ms │                                1313.07 ms │     no change │
│ QQuery 19 │   27.23 ms │                                  27.23 ms │     no change │
│ QQuery 20 │  512.54 ms │                                 516.26 ms │     no change │
│ QQuery 21 │  509.47 ms │                                 513.57 ms │     no change │
│ QQuery 22 │  989.47 ms │                                 985.77 ms │     no change │
│ QQuery 23 │ 3061.24 ms │                                3065.81 ms │     no change │
│ QQuery 24 │   41.89 ms │                                  40.89 ms │     no change │
│ QQuery 25 │  112.51 ms │                                 112.12 ms │     no change │
│ QQuery 26 │   41.32 ms │                                  43.42 ms │  1.05x slower │
│ QQuery 27 │  513.12 ms │                                 543.31 ms │  1.06x slower │
│ QQuery 28 │ 2928.70 ms │                                2913.50 ms │     no change │
│ QQuery 29 │   41.12 ms │                                  41.32 ms │     no change │
│ QQuery 30 │  305.74 ms │                                 309.89 ms │     no change │
│ QQuery 31 │  280.96 ms │                                 278.60 ms │     no change │
│ QQuery 32 │  913.90 ms │                                 969.33 ms │  1.06x slower │
│ QQuery 33 │ 1460.43 ms │                                1460.86 ms │     no change │
│ QQuery 34 │ 1458.29 ms │                                1460.09 ms │     no change │
│ QQuery 35 │  347.57 ms │                                 333.10 ms │     no change │
│ QQuery 36 │   72.00 ms │                                  67.07 ms │ +1.07x faster │
│ QQuery 37 │   38.43 ms │                                  35.87 ms │ +1.07x faster │
│ QQuery 38 │   44.03 ms │                                  40.77 ms │ +1.08x faster │
│ QQuery 39 │  166.63 ms │                                 135.36 ms │ +1.23x faster │
│ QQuery 40 │   15.84 ms │                                  15.09 ms │     no change │
│ QQuery 41 │   15.36 ms │                                  13.51 ms │ +1.14x faster │
│ QQuery 42 │   14.88 ms │                                  12.93 ms │ +1.15x faster │
└───────────┴────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 19128.84ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 19357.18ms │
│ Average Time (HEAD)                                      │   444.86ms │
│ Average Time (check-if-sorting-without-concat-is-better) │   450.17ms │
│ Queries Faster                                           │          6 │
│ Queries Slower                                           │          9 │
│ Queries with No Change                                   │         28 │
│ Queries with Failure                                     │          0 │
└──────────────────────────────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ check-if-sorting-without-concat-is-better ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.32 / 4.29 ±5.74 / 15.76 ms │              1.39 / 4.44 ±5.89 / 16.22 ms │     no change │
│ QQuery 1  │        12.03 / 12.58 ±0.29 / 12.88 ms │            12.19 / 12.65 ±0.28 / 13.05 ms │     no change │
│ QQuery 2  │        37.29 / 37.57 ±0.29 / 38.10 ms │            37.21 / 37.67 ±0.41 / 38.43 ms │     no change │
│ QQuery 3  │        30.65 / 31.96 ±1.11 / 33.30 ms │            32.35 / 32.85 ±0.40 / 33.49 ms │     no change │
│ QQuery 4  │     222.84 / 223.96 ±1.07 / 225.58 ms │         265.69 / 267.23 ±1.23 / 269.28 ms │  1.19x slower │
│ QQuery 5  │     275.33 / 276.91 ±1.71 / 279.95 ms │         305.32 / 310.39 ±2.96 / 313.80 ms │  1.12x slower │
│ QQuery 6  │           1.28 / 1.41 ±0.21 / 1.82 ms │               1.46 / 1.61 ±0.20 / 2.00 ms │  1.14x slower │
│ QQuery 7  │        13.12 / 13.32 ±0.13 / 13.52 ms │            13.41 / 13.61 ±0.20 / 13.97 ms │     no change │
│ QQuery 8  │    327.79 / 367.42 ±20.66 / 385.77 ms │        326.34 / 334.15 ±11.65 / 357.34 ms │ +1.10x faster │
│ QQuery 9  │    451.64 / 467.46 ±14.85 / 494.37 ms │        449.60 / 463.93 ±10.82 / 482.53 ms │     no change │
│ QQuery 10 │        70.83 / 71.85 ±0.74 / 72.76 ms │            68.94 / 71.79 ±2.66 / 75.76 ms │     no change │
│ QQuery 11 │        81.97 / 82.99 ±0.96 / 84.82 ms │            84.87 / 85.80 ±0.56 / 86.40 ms │     no change │
│ QQuery 12 │     269.07 / 273.79 ±6.25 / 285.71 ms │         296.11 / 306.11 ±6.61 / 315.71 ms │  1.12x slower │
│ QQuery 13 │    369.41 / 380.65 ±11.29 / 397.60 ms │        405.38 / 432.05 ±19.10 / 462.21 ms │  1.14x slower │
│ QQuery 14 │     284.39 / 291.94 ±6.29 / 300.28 ms │         283.92 / 289.93 ±7.22 / 301.64 ms │     no change │
│ QQuery 15 │    269.64 / 281.96 ±10.55 / 294.63 ms │        271.91 / 304.03 ±35.01 / 370.73 ms │  1.08x slower │
│ QQuery 16 │    607.64 / 624.57 ±21.51 / 666.40 ms │         615.34 / 620.51 ±5.84 / 631.12 ms │     no change │
│ QQuery 17 │    626.31 / 667.76 ±34.80 / 715.92 ms │        637.05 / 652.26 ±12.07 / 668.80 ms │     no change │
│ QQuery 18 │ 1263.61 / 1285.90 ±21.30 / 1317.24 ms │     1313.07 / 1344.75 ±37.58 / 1415.07 ms │     no change │
│ QQuery 19 │        27.23 / 27.47 ±0.24 / 27.93 ms │            27.23 / 29.02 ±2.21 / 33.20 ms │  1.06x slower │
│ QQuery 20 │    512.54 / 528.36 ±16.06 / 555.27 ms │        516.26 / 527.44 ±13.14 / 553.17 ms │     no change │
│ QQuery 21 │     509.47 / 518.20 ±4.64 / 522.55 ms │         513.57 / 520.92 ±7.05 / 531.15 ms │     no change │
│ QQuery 22 │  989.47 / 1021.87 ±26.39 / 1053.34 ms │      985.77 / 1006.57 ±19.66 / 1039.82 ms │     no change │
│ QQuery 23 │ 3061.24 / 3218.36 ±97.37 / 3326.40 ms │     3065.81 / 3108.65 ±35.32 / 3165.34 ms │     no change │
│ QQuery 24 │        41.89 / 44.93 ±4.96 / 54.78 ms │            40.89 / 44.04 ±4.62 / 53.14 ms │     no change │
│ QQuery 25 │    112.51 / 120.24 ±12.45 / 145.02 ms │         112.12 / 115.95 ±2.38 / 119.42 ms │     no change │
│ QQuery 26 │        41.32 / 43.14 ±1.44 / 45.51 ms │            43.42 / 45.13 ±1.17 / 46.49 ms │     no change │
│ QQuery 27 │     513.12 / 520.23 ±3.96 / 524.75 ms │         543.31 / 550.10 ±5.35 / 559.78 ms │  1.06x slower │
│ QQuery 28 │ 2928.70 / 2959.25 ±21.72 / 2988.82 ms │     2913.50 / 3021.86 ±74.57 / 3147.14 ms │     no change │
│ QQuery 29 │        41.12 / 43.17 ±3.17 / 49.33 ms │           41.32 / 53.79 ±14.99 / 74.41 ms │  1.25x slower │
│ QQuery 30 │     305.74 / 308.97 ±2.57 / 312.04 ms │        309.89 / 326.72 ±16.86 / 358.08 ms │  1.06x slower │
│ QQuery 31 │    280.96 / 295.97 ±11.98 / 317.03 ms │         278.60 / 288.34 ±5.50 / 294.47 ms │     no change │
│ QQuery 32 │  913.90 / 1003.66 ±70.25 / 1128.73 ms │       969.33 / 988.42 ±20.88 / 1017.89 ms │     no change │
│ QQuery 33 │ 1460.43 / 1498.57 ±25.20 / 1535.77 ms │     1460.86 / 1556.27 ±98.82 / 1690.78 ms │     no change │
│ QQuery 34 │ 1458.29 / 1545.84 ±51.77 / 1616.18 ms │     1460.09 / 1492.41 ±27.07 / 1542.12 ms │     no change │
│ QQuery 35 │     347.57 / 359.83 ±8.18 / 371.28 ms │        333.10 / 367.46 ±45.59 / 457.30 ms │     no change │
│ QQuery 36 │        72.00 / 78.58 ±5.43 / 84.99 ms │            67.07 / 73.32 ±4.38 / 80.32 ms │ +1.07x faster │
│ QQuery 37 │        38.43 / 45.49 ±5.62 / 53.27 ms │            35.87 / 42.00 ±4.52 / 48.76 ms │ +1.08x faster │
│ QQuery 38 │        44.03 / 46.73 ±1.74 / 48.54 ms │            40.77 / 43.35 ±2.88 / 48.14 ms │ +1.08x faster │
│ QQuery 39 │    166.63 / 181.52 ±14.85 / 204.59 ms │        135.36 / 155.47 ±13.80 / 174.78 ms │ +1.17x faster │
│ QQuery 40 │        15.84 / 16.37 ±0.63 / 17.60 ms │            15.09 / 15.78 ±0.63 / 16.61 ms │     no change │
│ QQuery 41 │        15.36 / 15.89 ±0.39 / 16.31 ms │            13.51 / 15.58 ±3.40 / 22.35 ms │     no change │
│ QQuery 42 │        14.88 / 15.07 ±0.18 / 15.38 ms │            12.93 / 13.22 ±0.19 / 13.46 ms │ +1.14x faster │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 19855.98ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 19987.54ms │
│ Average Time (HEAD)                                      │   461.77ms │
│ Average Time (check-if-sorting-without-concat-is-better) │   464.83ms │
│ Queries Faster                                           │          6 │
│ Queries Slower                                           │         10 │
│ Queries with No Change                                   │         27 │
│ Queries with Failure                                     │          0 │
└──────────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric Value
Wall time 100.0s
Peak memory 12.8 GiB
Avg memory 4.6 GiB
CPU user 1011.8s
CPU sys 74.8s
Peak spill 0 B

clickbench_partitioned — branch

Metric Value
Wall time 105.0s
Peak memory 11.3 GiB
Avg memory 4.1 GiB
CPU user 1016.6s
CPU sys 76.0s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark tpch
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃     HEAD ┃ check-if-sorting-without-concat-is-better ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 40.82 ms │                                  37.86 ms │ +1.08x faster │
│ QQuery 2  │ 20.27 ms │                                  18.63 ms │ +1.09x faster │
│ QQuery 3  │ 30.88 ms │                                  28.41 ms │ +1.09x faster │
│ QQuery 4  │ 18.86 ms │                                  17.55 ms │ +1.07x faster │
│ QQuery 5  │ 39.13 ms │                                  35.02 ms │ +1.12x faster │
│ QQuery 6  │ 17.18 ms │                                  15.88 ms │ +1.08x faster │
│ QQuery 7  │ 47.96 ms │                                  41.14 ms │ +1.17x faster │
│ QQuery 8  │ 41.56 ms │                                  40.78 ms │     no change │
│ QQuery 9  │ 49.26 ms │                                  49.32 ms │     no change │
│ QQuery 10 │ 41.81 ms │                                  41.85 ms │     no change │
│ QQuery 11 │ 13.42 ms │                                  13.22 ms │     no change │
│ QQuery 12 │ 23.87 ms │                                  23.67 ms │     no change │
│ QQuery 13 │ 39.87 ms │                                  39.03 ms │     no change │
│ QQuery 14 │ 25.67 ms │                                  24.29 ms │ +1.06x faster │
│ QQuery 15 │ 32.73 ms │                                  30.56 ms │ +1.07x faster │
│ QQuery 16 │ 14.76 ms │                                  13.71 ms │ +1.08x faster │
│ QQuery 17 │ 75.71 ms │                                  69.10 ms │ +1.10x faster │
│ QQuery 18 │ 59.54 ms │                                  57.45 ms │     no change │
│ QQuery 19 │ 33.34 ms │                                  32.27 ms │     no change │
│ QQuery 20 │ 30.97 ms │                                  30.98 ms │     no change │
│ QQuery 21 │ 55.31 ms │                                  54.59 ms │     no change │
│ QQuery 22 │ 13.85 ms │                                  13.72 ms │     no change │
└───────────┴──────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 766.76ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 729.02ms │
│ Average Time (HEAD)                                      │  34.85ms │
│ Average Time (check-if-sorting-without-concat-is-better) │  33.14ms │
│ Queries Faster                                           │       11 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       11 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃ check-if-sorting-without-concat-is-better ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 40.82 / 42.28 ±1.50 / 44.42 ms │            37.86 / 39.04 ±1.15 / 40.44 ms │ +1.08x faster │
│ QQuery 2  │ 20.27 / 20.53 ±0.28 / 21.02 ms │            18.63 / 18.85 ±0.28 / 19.38 ms │ +1.09x faster │
│ QQuery 3  │ 30.88 / 31.67 ±1.15 / 33.91 ms │            28.41 / 28.57 ±0.16 / 28.87 ms │ +1.11x faster │
│ QQuery 4  │ 18.86 / 19.34 ±0.38 / 20.04 ms │            17.55 / 17.67 ±0.10 / 17.84 ms │ +1.09x faster │
│ QQuery 5  │ 39.13 / 39.33 ±0.14 / 39.48 ms │            35.02 / 35.40 ±0.26 / 35.76 ms │ +1.11x faster │
│ QQuery 6  │ 17.18 / 17.85 ±1.03 / 19.89 ms │            15.88 / 16.05 ±0.13 / 16.27 ms │ +1.11x faster │
│ QQuery 7  │ 47.96 / 49.49 ±1.37 / 51.66 ms │            41.14 / 42.59 ±1.25 / 44.38 ms │ +1.16x faster │
│ QQuery 8  │ 41.56 / 44.65 ±2.71 / 47.73 ms │            40.78 / 42.67 ±2.22 / 46.81 ms │     no change │
│ QQuery 9  │ 49.26 / 49.97 ±0.61 / 51.02 ms │            49.32 / 50.44 ±0.97 / 52.13 ms │     no change │
│ QQuery 10 │ 41.81 / 42.03 ±0.18 / 42.32 ms │            41.85 / 42.00 ±0.08 / 42.07 ms │     no change │
│ QQuery 11 │ 13.42 / 13.55 ±0.14 / 13.79 ms │            13.22 / 13.41 ±0.26 / 13.92 ms │     no change │
│ QQuery 12 │ 23.87 / 24.61 ±1.20 / 26.97 ms │            23.67 / 23.99 ±0.34 / 24.60 ms │     no change │
│ QQuery 13 │ 39.87 / 42.17 ±2.59 / 46.47 ms │            39.03 / 40.56 ±2.11 / 44.57 ms │     no change │
│ QQuery 14 │ 25.67 / 25.84 ±0.16 / 26.15 ms │            24.29 / 24.42 ±0.13 / 24.66 ms │ +1.06x faster │
│ QQuery 15 │ 32.73 / 33.97 ±1.58 / 36.59 ms │            30.56 / 31.11 ±0.59 / 32.11 ms │ +1.09x faster │
│ QQuery 16 │ 14.76 / 14.85 ±0.05 / 14.92 ms │            13.71 / 14.02 ±0.17 / 14.25 ms │ +1.06x faster │
│ QQuery 17 │ 75.71 / 76.70 ±0.94 / 77.98 ms │            69.10 / 70.23 ±0.92 / 71.38 ms │ +1.09x faster │
│ QQuery 18 │ 59.54 / 62.16 ±1.83 / 63.74 ms │            57.45 / 60.08 ±1.82 / 62.63 ms │     no change │
│ QQuery 19 │ 33.34 / 33.59 ±0.28 / 34.11 ms │            32.27 / 32.64 ±0.38 / 33.35 ms │     no change │
│ QQuery 20 │ 30.97 / 31.62 ±0.48 / 32.19 ms │            30.98 / 31.42 ±0.44 / 32.26 ms │     no change │
│ QQuery 21 │ 55.31 / 56.19 ±0.57 / 57.10 ms │            54.59 / 55.06 ±0.25 / 55.28 ms │     no change │
│ QQuery 22 │ 13.85 / 14.07 ±0.21 / 14.40 ms │            13.72 / 13.93 ±0.18 / 14.26 ms │     no change │
└───────────┴────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 786.47ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 744.15ms │
│ Average Time (HEAD)                                      │  35.75ms │
│ Average Time (check-if-sorting-without-concat-is-better) │  33.82ms │
│ Queries Faster                                           │       11 │
│ Queries Slower                                           │        0 │
│ Queries with No Change                                   │       11 │
│ Queries with Failure                                     │        0 │
└──────────────────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 1.1 GiB
Avg memory 457.4 MiB
CPU user 22.2s
CPU sys 1.9s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.2 GiB
Avg memory 494.5 MiB
CPU user 20.7s
CPU sys 1.6s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark tpch10
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark tpch_sf10.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃      HEAD ┃ check-if-sorting-without-concat-is-better ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 308.06 ms │                                 306.36 ms │ no change │
│ QQuery 2  │  89.34 ms │                                  89.51 ms │ no change │
│ QQuery 3  │ 217.96 ms │                                 218.73 ms │ no change │
│ QQuery 4  │ 113.84 ms │                                 113.50 ms │ no change │
│ QQuery 5  │ 345.88 ms │                                 345.52 ms │ no change │
│ QQuery 6  │ 122.14 ms │                                 123.37 ms │ no change │
│ QQuery 7  │ 437.10 ms │                                 436.12 ms │ no change │
│ QQuery 8  │ 357.27 ms │                                 357.03 ms │ no change │
│ QQuery 9  │ 512.74 ms │                                 518.72 ms │ no change │
│ QQuery 10 │ 302.43 ms │                                 299.15 ms │ no change │
│ QQuery 11 │  63.32 ms │                                  63.11 ms │ no change │
│ QQuery 12 │ 179.02 ms │                                 176.44 ms │ no change │
│ QQuery 13 │ 297.33 ms │                                 289.19 ms │ no change │
│ QQuery 14 │ 169.80 ms │                                 170.61 ms │ no change │
│ QQuery 15 │ 297.20 ms │                                 299.84 ms │ no change │
│ QQuery 16 │  63.10 ms │                                  63.30 ms │ no change │
│ QQuery 17 │ 564.15 ms │                                 562.64 ms │ no change │
│ QQuery 18 │ 679.29 ms │                                 674.67 ms │ no change │
│ QQuery 19 │ 241.07 ms │                                 239.28 ms │ no change │
│ QQuery 20 │ 268.38 ms │                                 267.56 ms │ no change │
│ QQuery 21 │ 670.53 ms │                                 656.75 ms │ no change │
│ QQuery 22 │  60.87 ms │                                  58.21 ms │ no change │
└───────────┴───────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 6360.82ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 6329.60ms │
│ Average Time (HEAD)                                      │  289.13ms │
│ Average Time (check-if-sorting-without-concat-is-better) │  287.71ms │
│ Queries Faster                                           │         0 │
│ Queries Slower                                           │         0 │
│ Queries with No Change                                   │        22 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark tpch_sf10.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                               HEAD ┃ check-if-sorting-without-concat-is-better ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │  308.06 / 310.59 ±1.45 / 312.50 ms │         306.36 / 308.55 ±1.30 / 310.25 ms │     no change │
│ QQuery 2  │     89.34 / 90.60 ±0.98 / 92.05 ms │            89.51 / 92.42 ±1.77 / 94.36 ms │     no change │
│ QQuery 3  │  217.96 / 221.00 ±2.27 / 223.68 ms │         218.73 / 220.90 ±2.00 / 224.63 ms │     no change │
│ QQuery 4  │  113.84 / 115.20 ±1.37 / 117.11 ms │         113.50 / 115.14 ±1.33 / 117.13 ms │     no change │
│ QQuery 5  │  345.88 / 356.96 ±8.97 / 369.53 ms │         345.52 / 354.28 ±6.77 / 364.31 ms │     no change │
│ QQuery 6  │  122.14 / 124.73 ±2.24 / 128.85 ms │         123.37 / 125.23 ±2.24 / 129.64 ms │     no change │
│ QQuery 7  │  437.10 / 446.34 ±6.43 / 457.26 ms │        436.12 / 452.77 ±13.52 / 476.26 ms │     no change │
│ QQuery 8  │  357.27 / 360.96 ±2.14 / 363.30 ms │         357.03 / 361.21 ±3.81 / 368.31 ms │     no change │
│ QQuery 9  │ 512.74 / 526.91 ±12.07 / 544.33 ms │         518.72 / 523.15 ±4.39 / 530.56 ms │     no change │
│ QQuery 10 │  302.43 / 307.37 ±4.68 / 314.17 ms │        299.15 / 314.63 ±12.26 / 335.21 ms │     no change │
│ QQuery 11 │     63.32 / 63.86 ±0.38 / 64.36 ms │            63.11 / 64.71 ±1.01 / 66.19 ms │     no change │
│ QQuery 12 │  179.02 / 183.91 ±4.29 / 191.39 ms │         176.44 / 183.65 ±9.29 / 200.35 ms │     no change │
│ QQuery 13 │ 297.33 / 309.18 ±14.29 / 336.82 ms │        289.19 / 310.09 ±10.64 / 318.26 ms │     no change │
│ QQuery 14 │  169.80 / 173.76 ±4.86 / 183.23 ms │         170.61 / 174.60 ±3.65 / 179.32 ms │     no change │
│ QQuery 15 │  297.20 / 303.06 ±6.17 / 312.86 ms │         299.84 / 306.36 ±4.47 / 310.91 ms │     no change │
│ QQuery 16 │     63.10 / 65.10 ±1.23 / 66.56 ms │            63.30 / 65.45 ±2.13 / 69.28 ms │     no change │
│ QQuery 17 │  564.15 / 571.31 ±5.96 / 581.49 ms │         562.64 / 576.34 ±9.38 / 588.37 ms │     no change │
│ QQuery 18 │ 679.29 / 701.14 ±11.49 / 711.41 ms │        674.67 / 704.10 ±15.52 / 717.14 ms │     no change │
│ QQuery 19 │ 241.07 / 253.66 ±11.67 / 273.84 ms │        239.28 / 250.39 ±14.27 / 276.57 ms │     no change │
│ QQuery 20 │  268.38 / 276.09 ±6.52 / 284.77 ms │         267.56 / 276.16 ±4.88 / 281.82 ms │     no change │
│ QQuery 21 │  670.53 / 686.22 ±9.56 / 696.60 ms │         656.75 / 666.78 ±6.21 / 675.17 ms │     no change │
│ QQuery 22 │     60.87 / 64.18 ±3.62 / 71.05 ms │            58.21 / 60.49 ±2.50 / 65.12 ms │ +1.06x faster │
└───────────┴────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 6512.15ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 6507.40ms │
│ Average Time (HEAD)                                      │  296.01ms │
│ Average Time (check-if-sorting-without-concat-is-better) │  295.79ms │
│ Queries Faster                                           │         1 │
│ Queries Slower                                           │         0 │
│ Queries with No Change                                   │        21 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Resource Usage

tpch10 — base (merge-base)

Metric Value
Wall time 35.0s
Peak memory 4.7 GiB
Avg memory 1.6 GiB
CPU user 331.9s
CPU sys 18.9s
Peak spill 0 B

tpch10 — branch

Metric Value
Wall time 35.0s
Peak memory 4.8 GiB
Avg memory 1.5 GiB
CPU user 331.6s
CPU sys 19.1s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark tpcds
env:
  prefer_hash_join: "false"
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ check-if-sorting-without-concat-is-better ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.94 ms │                                   5.88 ms │     no change │
│ QQuery 2  │   80.91 ms │                                  80.78 ms │     no change │
│ QQuery 3  │   29.02 ms │                                  29.27 ms │     no change │
│ QQuery 4  │  476.76 ms │                                 478.04 ms │     no change │
│ QQuery 5  │   53.00 ms │                                  52.35 ms │     no change │
│ QQuery 6  │   36.74 ms │                                  35.86 ms │     no change │
│ QQuery 7  │   75.79 ms │                                  74.87 ms │     no change │
│ QQuery 8  │   37.03 ms │                                  36.76 ms │     no change │
│ QQuery 9  │   54.20 ms │                                  53.58 ms │     no change │
│ QQuery 10 │   64.23 ms │                                  62.69 ms │     no change │
│ QQuery 11 │  299.96 ms │                                 295.96 ms │     no change │
│ QQuery 12 │   29.51 ms │                                  29.14 ms │     no change │
│ QQuery 13 │  119.40 ms │                                 117.85 ms │     no change │
│ QQuery 14 │  421.16 ms │                                 422.25 ms │     no change │
│ QQuery 15 │   58.31 ms │                                  69.24 ms │  1.19x slower │
│ QQuery 16 │    6.96 ms │                                   8.06 ms │  1.16x slower │
│ QQuery 17 │   80.41 ms │                                  79.98 ms │     no change │
│ QQuery 18 │  105.22 ms │                                 104.74 ms │     no change │
│ QQuery 19 │   41.73 ms │                                  41.21 ms │     no change │
│ QQuery 20 │   35.65 ms │                                  35.78 ms │     no change │
│ QQuery 21 │   17.43 ms │                                  17.25 ms │     no change │
│ QQuery 22 │   64.07 ms │                                  63.18 ms │     no change │
│ QQuery 23 │  312.98 ms │                                 309.63 ms │     no change │
│ QQuery 24 │  195.62 ms │                                 192.40 ms │     no change │
│ QQuery 25 │  117.12 ms │                                 108.95 ms │ +1.08x faster │
│ QQuery 26 │   51.77 ms │                                  48.64 ms │ +1.06x faster │
│ QQuery 27 │    7.05 ms │                                   6.16 ms │ +1.14x faster │
│ QQuery 28 │   59.10 ms │                                  56.54 ms │     no change │
│ QQuery 29 │  101.29 ms │                                  96.54 ms │     no change │
│ QQuery 30 │   35.78 ms │                                  32.54 ms │ +1.10x faster │
│ QQuery 31 │  118.47 ms │                                 110.52 ms │ +1.07x faster │
│ QQuery 32 │   21.49 ms │                                  20.10 ms │ +1.07x faster │
│ QQuery 33 │   39.34 ms │                                  38.37 ms │     no change │
│ QQuery 34 │   10.69 ms │                                   9.94 ms │ +1.08x faster │
│ QQuery 35 │   76.59 ms │                                  72.37 ms │ +1.06x faster │
│ QQuery 36 │    6.63 ms │                                   6.08 ms │ +1.09x faster │
│ QQuery 37 │    7.61 ms │                                   6.91 ms │ +1.10x faster │
│ QQuery 38 │   69.51 ms │                                  61.63 ms │ +1.13x faster │
│ QQuery 39 │  104.16 ms │                                  89.09 ms │ +1.17x faster │
│ QQuery 40 │   26.95 ms │                                  23.68 ms │ +1.14x faster │
│ QQuery 41 │   12.55 ms │                                  11.07 ms │ +1.13x faster │
│ QQuery 42 │   25.50 ms │                                  23.67 ms │ +1.08x faster │
│ QQuery 43 │    6.15 ms │                                   5.38 ms │ +1.14x faster │
│ QQuery 44 │   10.76 ms │                                   9.59 ms │ +1.12x faster │
│ QQuery 45 │   49.88 ms │                                  38.63 ms │ +1.29x faster │
│ QQuery 46 │   13.18 ms │                                  12.27 ms │ +1.07x faster │
│ QQuery 47 │  279.76 ms │                                 222.67 ms │ +1.26x faster │
│ QQuery 48 │  100.33 ms │                                  95.45 ms │     no change │
│ QQuery 49 │   72.76 ms │                                  71.91 ms │     no change │
│ QQuery 50 │   59.09 ms │                                  58.24 ms │     no change │
│ QQuery 51 │   92.61 ms │                                  91.19 ms │     no change │
│ QQuery 52 │   24.98 ms │                                  23.80 ms │     no change │
│ QQuery 53 │   29.15 ms │                                  28.99 ms │     no change │
│ QQuery 54 │   54.38 ms │                                  54.23 ms │     no change │
│ QQuery 55 │   23.08 ms │                                  23.39 ms │     no change │
│ QQuery 56 │   39.21 ms │                                  39.06 ms │     no change │
│ QQuery 57 │  174.53 ms │                                 173.31 ms │     no change │
│ QQuery 58 │  111.66 ms │                                 111.64 ms │     no change │
│ QQuery 59 │  116.97 ms │                                 117.32 ms │     no change │
│ QQuery 60 │   39.76 ms │                                  39.57 ms │     no change │
│ QQuery 61 │   12.47 ms │                                  12.42 ms │     no change │
│ QQuery 62 │   46.20 ms │                                  45.78 ms │     no change │
│ QQuery 63 │   29.10 ms │                                  29.59 ms │     no change │
│ QQuery 64 │  364.99 ms │                                 393.91 ms │  1.08x slower │
│ QQuery 65 │  128.03 ms │                                 130.04 ms │     no change │
│ QQuery 66 │   88.01 ms │                                  87.95 ms │     no change │
│ QQuery 67 │  276.61 ms │                                 281.78 ms │     no change │
│ QQuery 68 │   13.07 ms │                                  13.63 ms │     no change │
│ QQuery 69 │   60.50 ms │                                  59.68 ms │     no change │
│ QQuery 70 │  112.25 ms │                                 113.39 ms │     no change │
│ QQuery 71 │   37.32 ms │                                  37.78 ms │     no change │
│ QQuery 72 │ 1822.29 ms │                                1932.11 ms │  1.06x slower │
│ QQuery 73 │   10.37 ms │                                  10.84 ms │     no change │
│ QQuery 74 │  171.18 ms │                                 207.35 ms │  1.21x slower │
│ QQuery 75 │  155.10 ms │                                 147.08 ms │ +1.05x faster │
│ QQuery 76 │   37.87 ms │                                  35.26 ms │ +1.07x faster │
│ QQuery 77 │   65.77 ms │                                  61.42 ms │ +1.07x faster │
│ QQuery 78 │  217.48 ms │                                 233.57 ms │  1.07x slower │
│ QQuery 79 │   66.17 ms │                                  66.14 ms │     no change │
│ QQuery 80 │  100.04 ms │                                  99.12 ms │     no change │
│ QQuery 81 │   25.72 ms │                                  26.43 ms │     no change │
│ QQuery 82 │   16.34 ms │                                  16.04 ms │     no change │
│ QQuery 83 │   34.49 ms │                                  34.01 ms │     no change │
│ QQuery 84 │   29.64 ms │                                  29.43 ms │     no change │
│ QQuery 85 │  102.03 ms │                                 102.44 ms │     no change │
│ QQuery 86 │   24.93 ms │                                  24.68 ms │     no change │
│ QQuery 87 │   63.19 ms │                                  61.81 ms │     no change │
│ QQuery 88 │   63.79 ms │                                  63.56 ms │     no change │
│ QQuery 89 │   35.44 ms │                                  35.28 ms │     no change │
│ QQuery 90 │   17.12 ms │                                  17.06 ms │     no change │
│ QQuery 91 │   45.51 ms │                                  45.29 ms │     no change │
│ QQuery 92 │   29.50 ms │                                  28.53 ms │     no change │
│ QQuery 93 │   50.16 ms │                                  50.03 ms │     no change │
│ QQuery 94 │   37.86 ms │                                  38.04 ms │     no change │
│ QQuery 95 │   79.98 ms │                                  80.85 ms │     no change │
│ QQuery 96 │   23.66 ms │                                  24.50 ms │     no change │
│ QQuery 97 │   52.32 ms │                                  53.65 ms │     no change │
│ QQuery 98 │   42.85 ms │                                  44.93 ms │     no change │
│ QQuery 99 │   69.99 ms │                                  71.00 ms │     no change │
└───────────┴────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 9551.24ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 9582.56ms │
│ Average Time (HEAD)                                      │   96.48ms │
│ Average Time (check-if-sorting-without-concat-is-better) │   96.79ms │
│ Queries Faster                                           │        23 │
│ Queries Slower                                           │         6 │
│ Queries with No Change                                   │        70 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ check-if-sorting-without-concat-is-better ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.94 / 6.48 ±0.95 / 8.37 ms │               5.88 / 6.39 ±0.91 / 8.21 ms │     no change │
│ QQuery 2  │        80.91 / 81.18 ±0.23 / 81.50 ms │            80.78 / 80.98 ±0.15 / 81.18 ms │     no change │
│ QQuery 3  │        29.02 / 29.33 ±0.23 / 29.71 ms │            29.27 / 29.60 ±0.24 / 30.00 ms │     no change │
│ QQuery 4  │     476.76 / 485.84 ±9.37 / 502.59 ms │         478.04 / 480.29 ±1.81 / 483.11 ms │     no change │
│ QQuery 5  │        53.00 / 54.12 ±1.02 / 55.65 ms │            52.35 / 52.68 ±0.26 / 52.99 ms │     no change │
│ QQuery 6  │        36.74 / 37.10 ±0.29 / 37.49 ms │            35.86 / 36.64 ±0.42 / 37.12 ms │     no change │
│ QQuery 7  │        75.79 / 77.05 ±0.90 / 78.16 ms │            74.87 / 75.19 ±0.48 / 76.13 ms │     no change │
│ QQuery 8  │        37.03 / 38.47 ±2.30 / 43.06 ms │            36.76 / 36.90 ±0.14 / 37.12 ms │     no change │
│ QQuery 9  │        54.20 / 55.84 ±1.14 / 57.78 ms │            53.58 / 54.63 ±0.90 / 55.78 ms │     no change │
│ QQuery 10 │        64.23 / 64.57 ±0.26 / 64.88 ms │            62.69 / 63.13 ±0.29 / 63.53 ms │     no change │
│ QQuery 11 │     299.96 / 302.47 ±1.39 / 304.22 ms │         295.96 / 300.20 ±2.19 / 302.19 ms │     no change │
│ QQuery 12 │        29.51 / 29.76 ±0.30 / 30.33 ms │            29.14 / 29.34 ±0.17 / 29.60 ms │     no change │
│ QQuery 13 │     119.40 / 120.27 ±0.82 / 121.78 ms │         117.85 / 118.77 ±1.02 / 120.72 ms │     no change │
│ QQuery 14 │     421.16 / 425.34 ±4.23 / 433.39 ms │         422.25 / 433.23 ±8.99 / 442.92 ms │     no change │
│ QQuery 15 │        58.31 / 58.58 ±0.20 / 58.82 ms │            69.24 / 70.76 ±1.26 / 72.60 ms │  1.21x slower │
│ QQuery 16 │           6.96 / 7.16 ±0.29 / 7.74 ms │               8.06 / 8.40 ±0.54 / 9.48 ms │  1.17x slower │
│ QQuery 17 │        80.41 / 81.58 ±1.25 / 83.63 ms │            79.98 / 83.11 ±3.40 / 87.43 ms │     no change │
│ QQuery 18 │     105.22 / 107.28 ±3.02 / 113.20 ms │         104.74 / 106.04 ±1.19 / 108.02 ms │     no change │
│ QQuery 19 │        41.73 / 41.88 ±0.12 / 42.00 ms │            41.21 / 42.24 ±1.16 / 44.48 ms │     no change │
│ QQuery 20 │        35.65 / 36.23 ±0.36 / 36.63 ms │            35.78 / 36.16 ±0.37 / 36.82 ms │     no change │
│ QQuery 21 │        17.43 / 17.74 ±0.27 / 18.19 ms │            17.25 / 17.52 ±0.17 / 17.74 ms │     no change │
│ QQuery 22 │        64.07 / 64.66 ±0.65 / 65.92 ms │            63.18 / 67.15 ±3.12 / 70.96 ms │     no change │
│ QQuery 23 │     312.98 / 314.09 ±0.90 / 315.45 ms │         309.63 / 316.66 ±6.56 / 327.86 ms │     no change │
│ QQuery 24 │     195.62 / 204.58 ±9.07 / 221.01 ms │         192.40 / 196.24 ±4.53 / 204.69 ms │     no change │
│ QQuery 25 │     117.12 / 119.85 ±2.98 / 125.56 ms │         108.95 / 110.78 ±1.23 / 112.09 ms │ +1.08x faster │
│ QQuery 26 │        51.77 / 52.18 ±0.32 / 52.74 ms │            48.64 / 48.87 ±0.16 / 49.14 ms │ +1.07x faster │
│ QQuery 27 │           7.05 / 7.14 ±0.11 / 7.33 ms │               6.16 / 6.36 ±0.16 / 6.65 ms │ +1.12x faster │
│ QQuery 28 │        59.10 / 60.80 ±1.65 / 63.84 ms │            56.54 / 58.57 ±2.30 / 61.62 ms │     no change │
│ QQuery 29 │     101.29 / 102.33 ±0.71 / 103.49 ms │           96.54 / 98.58 ±2.07 / 101.61 ms │     no change │
│ QQuery 30 │        35.78 / 35.96 ±0.16 / 36.24 ms │            32.54 / 32.75 ±0.22 / 33.05 ms │ +1.10x faster │
│ QQuery 31 │     118.47 / 120.01 ±2.15 / 124.28 ms │         110.52 / 112.04 ±1.68 / 115.22 ms │ +1.07x faster │
│ QQuery 32 │        21.49 / 22.22 ±0.38 / 22.56 ms │            20.10 / 20.55 ±0.35 / 21.17 ms │ +1.08x faster │
│ QQuery 33 │        39.34 / 39.64 ±0.28 / 40.14 ms │            38.37 / 38.78 ±0.37 / 39.44 ms │     no change │
│ QQuery 34 │        10.69 / 11.09 ±0.28 / 11.57 ms │             9.94 / 10.20 ±0.31 / 10.79 ms │ +1.09x faster │
│ QQuery 35 │        76.59 / 80.29 ±2.79 / 83.71 ms │            72.37 / 73.80 ±2.13 / 78.04 ms │ +1.09x faster │
│ QQuery 36 │           6.63 / 6.77 ±0.13 / 7.01 ms │               6.08 / 6.21 ±0.14 / 6.48 ms │ +1.09x faster │
│ QQuery 37 │           7.61 / 7.79 ±0.11 / 7.97 ms │               6.91 / 6.98 ±0.08 / 7.11 ms │ +1.12x faster │
│ QQuery 38 │        69.51 / 70.11 ±0.65 / 71.24 ms │            61.63 / 62.20 ±0.52 / 63.13 ms │ +1.13x faster │
│ QQuery 39 │     104.16 / 105.44 ±1.18 / 107.49 ms │            89.09 / 89.95 ±0.68 / 90.79 ms │ +1.17x faster │
│ QQuery 40 │        26.95 / 27.86 ±0.72 / 28.74 ms │            23.68 / 24.46 ±0.62 / 25.49 ms │ +1.14x faster │
│ QQuery 41 │        12.55 / 13.12 ±0.77 / 14.63 ms │            11.07 / 11.28 ±0.26 / 11.79 ms │ +1.16x faster │
│ QQuery 42 │        25.50 / 26.10 ±0.41 / 26.59 ms │            23.67 / 23.97 ±0.36 / 24.58 ms │ +1.09x faster │
│ QQuery 43 │           6.15 / 6.20 ±0.06 / 6.32 ms │               5.38 / 5.53 ±0.19 / 5.91 ms │ +1.12x faster │
│ QQuery 44 │        10.76 / 10.87 ±0.08 / 11.02 ms │               9.59 / 9.71 ±0.08 / 9.82 ms │ +1.12x faster │
│ QQuery 45 │        49.88 / 50.79 ±0.95 / 52.58 ms │            38.63 / 39.40 ±0.61 / 40.50 ms │ +1.29x faster │
│ QQuery 46 │        13.18 / 13.67 ±0.44 / 14.40 ms │            12.27 / 12.38 ±0.10 / 12.53 ms │ +1.10x faster │
│ QQuery 47 │     279.76 / 286.33 ±4.25 / 290.23 ms │         222.67 / 226.71 ±2.91 / 230.19 ms │ +1.26x faster │
│ QQuery 48 │     100.33 / 101.70 ±1.26 / 103.51 ms │            95.45 / 96.21 ±0.47 / 96.82 ms │ +1.06x faster │
│ QQuery 49 │        72.76 / 77.64 ±5.82 / 88.93 ms │            71.91 / 72.35 ±0.56 / 73.43 ms │ +1.07x faster │
│ QQuery 50 │        59.09 / 59.58 ±0.41 / 60.32 ms │            58.24 / 58.97 ±0.42 / 59.56 ms │     no change │
│ QQuery 51 │        92.61 / 94.59 ±1.33 / 96.68 ms │            91.19 / 94.38 ±2.08 / 97.40 ms │     no change │
│ QQuery 52 │        24.98 / 27.06 ±2.71 / 32.33 ms │            23.80 / 24.05 ±0.21 / 24.30 ms │ +1.12x faster │
│ QQuery 53 │        29.15 / 29.51 ±0.31 / 30.02 ms │            28.99 / 29.24 ±0.24 / 29.56 ms │     no change │
│ QQuery 54 │        54.38 / 54.99 ±0.50 / 55.67 ms │            54.23 / 54.73 ±0.30 / 55.04 ms │     no change │
│ QQuery 55 │        23.08 / 23.46 ±0.23 / 23.81 ms │            23.39 / 24.76 ±2.15 / 28.99 ms │  1.06x slower │
│ QQuery 56 │        39.21 / 39.41 ±0.11 / 39.55 ms │            39.06 / 39.86 ±0.51 / 40.49 ms │     no change │
│ QQuery 57 │     174.53 / 176.31 ±1.66 / 179.38 ms │         173.31 / 175.73 ±1.65 / 177.56 ms │     no change │
│ QQuery 58 │     111.66 / 113.64 ±1.85 / 116.85 ms │         111.64 / 112.81 ±0.89 / 114.27 ms │     no change │
│ QQuery 59 │     116.97 / 117.62 ±0.44 / 118.27 ms │         117.32 / 118.52 ±1.23 / 120.81 ms │     no change │
│ QQuery 60 │        39.76 / 40.61 ±1.10 / 42.75 ms │            39.57 / 40.36 ±0.94 / 42.18 ms │     no change │
│ QQuery 61 │        12.47 / 12.89 ±0.44 / 13.58 ms │            12.42 / 12.60 ±0.19 / 12.96 ms │     no change │
│ QQuery 62 │        46.20 / 46.83 ±1.08 / 48.98 ms │            45.78 / 46.14 ±0.21 / 46.40 ms │     no change │
│ QQuery 63 │        29.10 / 29.23 ±0.13 / 29.48 ms │            29.59 / 30.24 ±0.58 / 31.13 ms │     no change │
│ QQuery 64 │    364.99 / 374.96 ±11.99 / 398.38 ms │         393.91 / 397.64 ±2.89 / 402.08 ms │  1.06x slower │
│ QQuery 65 │     128.03 / 130.10 ±1.50 / 131.70 ms │         130.04 / 134.59 ±2.66 / 137.66 ms │     no change │
│ QQuery 66 │        88.01 / 90.19 ±3.88 / 97.93 ms │            87.95 / 88.54 ±0.60 / 89.38 ms │     no change │
│ QQuery 67 │     276.61 / 281.75 ±3.58 / 287.22 ms │         281.78 / 286.18 ±3.97 / 293.64 ms │     no change │
│ QQuery 68 │        13.07 / 13.27 ±0.13 / 13.46 ms │            13.63 / 14.22 ±0.84 / 15.88 ms │  1.07x slower │
│ QQuery 69 │        60.50 / 63.93 ±4.41 / 72.35 ms │            59.68 / 60.20 ±0.37 / 60.77 ms │ +1.06x faster │
│ QQuery 70 │     112.25 / 115.09 ±3.60 / 122.03 ms │         113.39 / 117.09 ±3.23 / 122.35 ms │     no change │
│ QQuery 71 │        37.32 / 37.86 ±0.28 / 38.12 ms │            37.78 / 38.77 ±0.59 / 39.37 ms │     no change │
│ QQuery 72 │ 1822.29 / 1922.04 ±70.64 / 2007.52 ms │     1932.11 / 2002.92 ±64.63 / 2091.61 ms │     no change │
│ QQuery 73 │        10.37 / 11.04 ±0.53 / 11.66 ms │            10.84 / 11.22 ±0.36 / 11.81 ms │     no change │
│ QQuery 74 │     171.18 / 174.28 ±1.85 / 176.20 ms │         207.35 / 210.33 ±2.44 / 214.01 ms │  1.21x slower │
│ QQuery 75 │     155.10 / 158.37 ±3.42 / 163.17 ms │         147.08 / 153.92 ±6.92 / 165.23 ms │     no change │
│ QQuery 76 │        37.87 / 38.34 ±0.47 / 39.15 ms │            35.26 / 36.37 ±1.75 / 39.86 ms │ +1.05x faster │
│ QQuery 77 │        65.77 / 66.39 ±0.51 / 67.10 ms │            61.42 / 61.66 ±0.16 / 61.90 ms │ +1.08x faster │
│ QQuery 78 │    217.48 / 228.57 ±13.30 / 254.65 ms │         233.57 / 236.78 ±2.75 / 241.67 ms │     no change │
│ QQuery 79 │        66.17 / 67.49 ±1.46 / 70.16 ms │            66.14 / 66.81 ±0.48 / 67.54 ms │     no change │
│ QQuery 80 │     100.04 / 104.17 ±2.67 / 107.65 ms │          99.12 / 100.81 ±2.50 / 105.78 ms │     no change │
│ QQuery 81 │        25.72 / 25.87 ±0.12 / 26.07 ms │            26.43 / 27.71 ±2.05 / 31.80 ms │  1.07x slower │
│ QQuery 82 │        16.34 / 16.53 ±0.18 / 16.84 ms │            16.04 / 16.45 ±0.30 / 16.80 ms │     no change │
│ QQuery 83 │        34.49 / 37.72 ±4.30 / 45.47 ms │            34.01 / 34.21 ±0.19 / 34.52 ms │ +1.10x faster │
│ QQuery 84 │        29.64 / 29.91 ±0.18 / 30.11 ms │            29.43 / 29.70 ±0.24 / 30.10 ms │     no change │
│ QQuery 85 │     102.03 / 103.34 ±0.99 / 104.95 ms │         102.44 / 105.14 ±2.59 / 108.65 ms │     no change │
│ QQuery 86 │        24.93 / 25.25 ±0.19 / 25.51 ms │            24.68 / 25.55 ±0.51 / 26.28 ms │     no change │
│ QQuery 87 │        63.19 / 65.10 ±2.45 / 69.91 ms │            61.81 / 62.13 ±0.25 / 62.56 ms │     no change │
│ QQuery 88 │        63.79 / 64.20 ±0.39 / 64.92 ms │            63.56 / 64.61 ±0.91 / 66.32 ms │     no change │
│ QQuery 89 │        35.44 / 35.77 ±0.27 / 36.13 ms │            35.28 / 36.01 ±0.57 / 36.86 ms │     no change │
│ QQuery 90 │        17.12 / 17.36 ±0.15 / 17.56 ms │            17.06 / 19.31 ±4.10 / 27.51 ms │  1.11x slower │
│ QQuery 91 │        45.51 / 46.72 ±1.55 / 49.77 ms │            45.29 / 45.87 ±0.56 / 46.94 ms │     no change │
│ QQuery 92 │        29.50 / 30.49 ±0.89 / 32.10 ms │            28.53 / 29.54 ±0.72 / 30.44 ms │     no change │
│ QQuery 93 │        50.16 / 51.37 ±1.08 / 52.79 ms │            50.03 / 50.80 ±0.51 / 51.62 ms │     no change │
│ QQuery 94 │        37.86 / 38.08 ±0.12 / 38.23 ms │            38.04 / 38.63 ±0.36 / 39.03 ms │     no change │
│ QQuery 95 │        79.98 / 81.01 ±0.70 / 81.82 ms │            80.85 / 82.81 ±1.40 / 84.54 ms │     no change │
│ QQuery 96 │        23.66 / 24.74 ±1.28 / 27.25 ms │            24.50 / 24.62 ±0.11 / 24.83 ms │     no change │
│ QQuery 97 │        52.32 / 53.44 ±0.87 / 54.84 ms │            53.65 / 54.12 ±0.74 / 55.58 ms │     no change │
│ QQuery 98 │        42.85 / 43.57 ±0.67 / 44.79 ms │            44.93 / 45.57 ±0.50 / 46.25 ms │     no change │
│ QQuery 99 │        69.99 / 71.79 ±2.62 / 76.97 ms │            71.00 / 73.13 ±2.47 / 77.95 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 9803.32ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 9787.23ms │
│ Average Time (HEAD)                                      │   99.02ms │
│ Average Time (check-if-sorting-without-concat-is-better) │   98.86ms │
│ Queries Faster                                           │        27 │
│ Queries Slower                                           │         8 │
│ Queries with No Change                                   │        64 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 50.0s
Peak memory 2.0 GiB
Avg memory 1.4 GiB
CPU user 215.0s
CPU sys 5.7s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 50.0s
Peak memory 2.1 GiB
Avg memory 1.5 GiB
CPU user 216.2s
CPU sys 5.5s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark sort_tpch
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark sort_tpch1.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query ┃      HEAD ┃ check-if-sorting-without-concat-is-better ┃    Change ┃
┡━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Q1    │ 124.83 ms │                                 125.54 ms │ no change │
│ Q2    │ 105.55 ms │                                 105.17 ms │ no change │
│ Q3    │ 638.34 ms │                                 634.53 ms │ no change │
│ Q4    │ 198.12 ms │                                 196.62 ms │ no change │
│ Q5    │ 261.97 ms │                                 263.80 ms │ no change │
│ Q6    │ 276.84 ms │                                 278.61 ms │ no change │
│ Q7    │ 447.06 ms │                                 444.38 ms │ no change │
│ Q8    │ 312.95 ms │                                 311.93 ms │ no change │
│ Q9    │ 337.26 ms │                                 334.45 ms │ no change │
│ Q10   │ 499.58 ms │                                 489.51 ms │ no change │
│ Q11   │ 246.06 ms │                                 244.94 ms │ no change │
└───────┴───────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 3448.58ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 3429.49ms │
│ Average Time (HEAD)                                      │  313.51ms │
│ Average Time (check-if-sorting-without-concat-is-better) │  311.77ms │
│ Queries Faster                                           │         0 │
│ Queries Slower                                           │         0 │
│ Queries with No Change                                   │        11 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark sort_tpch1.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query ┃                               HEAD ┃ check-if-sorting-without-concat-is-better ┃        Change ┃
┡━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ Q1    │  124.83 / 125.83 ±0.90 / 127.52 ms │         125.54 / 126.79 ±0.87 / 128.18 ms │     no change │
│ Q2    │  105.55 / 107.25 ±1.50 / 109.82 ms │         105.17 / 106.53 ±1.51 / 108.93 ms │     no change │
│ Q3    │  638.34 / 644.19 ±6.07 / 655.85 ms │         634.53 / 641.37 ±5.19 / 649.21 ms │     no change │
│ Q4    │  198.12 / 200.71 ±2.65 / 205.80 ms │         196.62 / 200.52 ±2.38 / 203.03 ms │     no change │
│ Q5    │  261.97 / 263.38 ±1.26 / 265.44 ms │         263.80 / 265.47 ±2.51 / 270.42 ms │     no change │
│ Q6    │  276.84 / 277.55 ±0.76 / 278.94 ms │         278.61 / 280.31 ±1.34 / 281.49 ms │     no change │
│ Q7    │  447.06 / 450.15 ±4.11 / 457.99 ms │         444.38 / 447.37 ±2.60 / 450.91 ms │     no change │
│ Q8    │  312.95 / 318.87 ±5.16 / 328.30 ms │        311.93 / 321.67 ±11.80 / 343.47 ms │     no change │
│ Q9    │  337.26 / 346.75 ±9.46 / 359.20 ms │        334.45 / 362.90 ±25.64 / 403.44 ms │     no change │
│ Q10   │ 499.58 / 527.49 ±16.13 / 543.64 ms │         489.51 / 492.99 ±2.43 / 496.63 ms │ +1.07x faster │
│ Q11   │ 246.06 / 258.56 ±14.67 / 284.26 ms │        244.94 / 265.86 ±26.93 / 312.02 ms │     no change │
└───────┴────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 3520.73ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 3511.78ms │
│ Average Time (HEAD)                                      │  320.07ms │
│ Average Time (check-if-sorting-without-concat-is-better) │  319.25ms │
│ Queries Faster                                           │         1 │
│ Queries Slower                                           │         0 │
│ Queries with No Change                                   │        10 │
│ Queries with Failure                                     │         0 │
└──────────────────────────────────────────────────────────┴───────────┘

Resource Usage

sort_tpch — base (merge-base)

Metric Value
Wall time 20.0s
Peak memory 2.2 GiB
Avg memory 1.1 GiB
CPU user 63.8s
CPU sys 3.2s
Peak spill 0 B

sort_tpch — branch

Metric Value
Wall time 20.0s
Peak memory 2.5 GiB
Avg memory 1.1 GiB
CPU user 63.5s
CPU sys 3.1s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Sep 9, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing check-if-sorting-without-concat-is-better (0dab556) to a5c809f (merge-base) diff

Run configuration 
run benchmark sort_tpch10
CPU Details (lscpu) 
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details 

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark sort_tpch10.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query ┃       HEAD ┃ check-if-sorting-without-concat-is-better ┃    Change ┃
┡━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Q1    │ 1661.01 ms │                                1670.00 ms │ no change │
│ Q2    │ 1425.53 ms │                                1425.07 ms │ no change │
│ Q3    │ 8052.67 ms │                                8107.89 ms │ no change │
│ Q4    │ 2873.43 ms │                                2869.38 ms │ no change │
│ Q5    │ 2650.56 ms │                                2669.73 ms │ no change │
│ Q6    │ 2791.80 ms │                                2780.45 ms │ no change │
│ Q7    │ 5125.07 ms │                                5113.20 ms │ no change │
│ Q8    │ 3986.70 ms │                                4041.66 ms │ no change │
│ Q9    │ 4484.13 ms │                                4504.06 ms │ no change │
│ Q10   │ 6796.66 ms │                                6791.56 ms │ no change │
│ Q11   │ 3358.13 ms │                                3344.84 ms │ no change │
└───────┴────────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 43205.70ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 43317.84ms │
│ Average Time (HEAD)                                      │  3927.79ms │
│ Average Time (check-if-sorting-without-concat-is-better) │  3937.99ms │
│ Queries Faster                                           │          0 │
│ Queries Slower                                           │          0 │
│ Queries with No Change                                   │         11 │
│ Queries with Failure                                     │          0 │
└──────────────────────────────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and check-if-sorting-without-concat-is-better
--------------------
Benchmark sort_tpch10.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query ┃                                   HEAD ┃ check-if-sorting-without-concat-is-better ┃    Change ┃
┡━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Q1    │   1661.01 / 1664.55 ±3.91 / 1672.18 ms │      1670.00 / 1675.34 ±2.94 / 1677.93 ms │ no change │
│ Q2    │  1425.53 / 1477.48 ±33.28 / 1510.74 ms │     1425.07 / 1446.00 ±13.63 / 1461.00 ms │ no change │
│ Q3    │  8052.67 / 8099.39 ±37.69 / 8165.44 ms │     8107.89 / 8209.82 ±53.87 / 8267.96 ms │ no change │
│ Q4    │  2873.43 / 2916.09 ±33.27 / 2959.07 ms │     2869.38 / 2900.97 ±22.06 / 2933.75 ms │ no change │
│ Q5    │  2650.56 / 2665.39 ±12.73 / 2687.10 ms │     2669.73 / 2690.07 ±26.44 / 2739.12 ms │ no change │
│ Q6    │  2791.80 / 2806.66 ±12.23 / 2822.19 ms │     2780.45 / 2808.95 ±19.02 / 2834.09 ms │ no change │
│ Q7    │  5125.07 / 5139.04 ±13.04 / 5156.53 ms │     5113.20 / 5216.51 ±94.22 / 5389.69 ms │ no change │
│ Q8    │  3986.70 / 4146.53 ±82.21 / 4208.92 ms │     4041.66 / 4158.19 ±92.59 / 4322.15 ms │ no change │
│ Q9    │  4484.13 / 4511.63 ±22.43 / 4549.75 ms │     4504.06 / 4532.89 ±24.76 / 4572.90 ms │ no change │
│ Q10   │ 6796.66 / 7104.38 ±173.80 / 7284.28 ms │    6791.56 / 7139.09 ±214.82 / 7399.23 ms │ no change │
│ Q11   │  3358.13 / 3404.82 ±50.41 / 3499.83 ms │     3344.84 / 3403.50 ±79.05 / 3558.16 ms │ no change │
└───────┴────────────────────────────────────────┴───────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                        │ 43935.97ms │
│ Total Time (check-if-sorting-without-concat-is-better)   │ 44181.32ms │
│ Average Time (HEAD)                                      │  3994.18ms │
│ Average Time (check-if-sorting-without-concat-is-better) │  4016.48ms │
│ Queries Faster                                           │          0 │
│ Queries Slower                                           │          0 │
│ Queries with No Change                                   │         11 │
│ Queries with Failure                                     │          0 │
└──────────────────────────────────────────────────────────┴────────────┘

Resource Usage

sort_tpch10 — base (merge-base)

Metric Value
Wall time 225.1s
Peak memory 19.2 GiB
Avg memory 7.2 GiB
CPU user 874.0s
CPU sys 72.5s
Peak spill 0 B

sort_tpch10 — branch

Metric Value
Wall time 225.1s
Peak memory 19.3 GiB
Avg memory 7.3 GiB
CPU user 883.1s
CPU sys 71.4s
Peak spill 0 B

File an issue against this benchmark runner

@rluvaton

rluvaton commented Sep 9, 2026

Copy link
Copy Markdown
Member Author

ok, this improve perf, but need to be in arrow

@rluvaton rluvaton closed this Sep 9, 2026
@rluvaton
rluvaton deleted the check-if-sorting-without-concat-is-better branch September 9, 2026 18:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

physical-plan Changes to the physical-plan crate

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@rluvaton @adriangbot