session: disable AMReX arena preallocation to fit T4 with EB#296
Merged
Conversation
AMReX's default GPU arena preallocates a large chunk of VRAM upfront (~all available memory minus headroom). With EB enabled, the metadata allocations on top of this push the T4 (15 GB) over the limit, OOM-ing at AMReX init. Pass amrex.the_arena_init_size=0 to use lazy allocation instead — each kernel allocates what it needs, when it needs it. Small runtime overhead but unblocks Colab T4 users.
Performance Benchmark Results
Fastest solver: gmres at 64³ (0.3892s) Benchmark: uniform block (analytical τ = (N-1)/N) |
Code Coverage ReportGenerated by CI — coverage data from gcovr |
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
AMReX's default GPU arena preallocates a large chunk of VRAM upfront (~all available memory minus headroom). With EB enabled, the metadata allocations on top of this push the T4 (15 GB) over the limit, OOM-ing at AMReX init. Pass amrex.the_arena_init_size=0 to use lazy allocation instead — each kernel allocates what it needs, when it needs it. Small runtime overhead but unblocks Colab T4 users.