forked from NoopurM/Hybrid_Node_Scheduler
-
Notifications
You must be signed in to change notification settings - Fork 0
rahulrane50/Hybrid_Node_Scheduler
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
Noopur Maheshwari - 111464061
Rahul Rane - 111465246
--------------------------------------------------------------------------------------------------------
IMPLEMENTATION:
Merge Sort:
fork-join implementation - gpu_merge_sort/merge_sort.cu
gpu_merge_sort/merge_sort.h
Matrix Multiplication:
fork-join implementation - gpu_mm/mm_rec.cu
gpu_mm/mm_rec.h
Scheduler:
Work stealing implementation - gpu_merge_sort/scheduler.cu
gpu_merge_sort/scheduler.h
gpu_mm/scheduler.cu
gpu_mm/scheduler.h
Threadpool:
CPU and GPU worker implementation - gpu_merge_sort/threadpool.cu
gpu_mm/threadpool.cu
Naive Bayes:
Bayes algo implementation - gpu_merge_sort/bayes.cu
gpu_merge_sort/bayes.h
gpu_merge_sort/training.data
gpu_mm/bayes.cu
gpu_mm/bayes.h
gpu_mm/training.data
--------------------------------------------------------------------------------------------------------
COMPILATION AND RUNNING:
We have provided Makefile in both folders which creates a single binary:
For ex.,
cd gpu_merge_sort
make
./merge_sort
All parameters are configurable and defined as macro in main files.
About
Hybrid Node Scheduler - A dynamic scheduler that schedules tasks and recursive fork-join jobs across CPUs and GPUs using work stealing algorithm. Tested on COMET Supercomputer.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published
Languages
- Cuda 82.0%
- C++ 16.5%
- Other 1.5%