Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add failed EBT experiment #172

Draft
wants to merge 1 commit into
base: master
Choose a base branch
from
Draft

Add failed EBT experiment #172

wants to merge 1 commit into from

Conversation

arj03
Copy link
Member

@arj03 arj03 commented Feb 12, 2021

Yesterday I tried a thing. I noticed that EBT is one of the slowest indexes, and looking at it, it's pretty clear why, it basically has to store everything message in level. On my machine EBT is 12 seconds to index from scratch. I was wondering if I could use some of the same things as I used for ssb-friends on this index and tried that. This takes the full index time down to 7 seconds, but it adds a startup overhead of 700ms, so I don't think its worth it. It does reduce the size of the database from 24mb to 1.4mb.

So this PR is mainly a learning thing and can be closed.

@github-actions
Copy link

Benchmark results

Part Duration
Migrate (+db1) 8549ms
Migrate (alone) 4159ms
Memory usage without indexes 514.49 MB = 26.17 MB + etc
Migrate (+db1 +db2) 7324ms
Migrate (+db2) 3589ms
Migrate continuation (+db2) 1696ms
Initial indexing 1260ms
Initial indexing maxCpu=86 5199ms
Initial indexing compat 365ms
Two indexes updating concurrently 1107ms
ssb-threads and ssb-friends 2662ms
ssb-threads and ssb-friends again 70ms
key one initial 115ms
key two 0ms
key one again 0ms
latest root posts 508ms
latest posts 5ms
votes one initial 576ms
votes again 0ms
hasRoot 7ms
hasRoot again 0ms
author one posts 345ms
author two posts 91ms
dedicated author one posts 331ms
dedicated author one posts again 0ms
Maximum memory usage 822.24 MB = 123.06 MB + etc

@staltz
Copy link
Member

staltz commented Feb 12, 2021

Ok, before closing, let me take a look at it to understand, and maybe we find another insight.

@staltz
Copy link
Member

staltz commented Feb 22, 2021

How about we mark this PR as a draft to signal that it's not supposed to be merged? (you can undo the draft state if you disagree)

@staltz staltz marked this pull request as draft February 22, 2021 09:50
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants