NVIDIA / NVIDIA/AMGX

Parallel solves

Open
#151 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Cuda
Stars
692
Forks
197
PR merge metrics
No merged PRs in 30d

Description

I wanted to know whether it is possible to run several solves in parallel, if they will all fit in GPU memory.

The use case is as follows - there are around 40 solves, each one of them takes around 10 seconds. I have been using a custom CG solver written in CUDA, where I could run e.g. 10 such solves in parallel.

After switching to AMGX, I am receiving a lot of exceptions, when several solves are trying to be run from different threads. I tried to perform the initialisation on single thread, but the result is the same.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No files, tests, or entry points are named. Start by reproducing the reported case of several AMGX solves running from different threads, including single-thread initialization, and capture the exceptions; done means determining whether concurrent solves are supported when GPU memory allows and documenting the cause or required usage.

Written by the indexing model from the issue text.

Assessment

Domain
hpc, performance
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
15/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.