Add CUDA async memory resource as an option #3447
Merged
Conversation
This file contains 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
Signed-off-by: Rong Ou <rong.ou@gmail.com>
rongou
added
feature request
abellina
previously approved these changes
Sep 10, 2021
jlowe
added
the
cudf_dependency
jlowe
reviewed
Sep 10, 2021
| "as \"DEFAULT\" is known to be unstable (https://github.com/NVIDIA/spark-rapids/issues/1141)") | ||
| .doc("Select the RMM pooling allocator to use. Valid values are \"DEFAULT\", \"ARENA\", " + | ||
| "\"ASYNC\", and \"NONE\". With \"DEFAULT\", `rmm::mr::pool_memory_resource` is used; with " + | ||
| "\"ARENA\", `rmm::mr::arena_memory_resource` is used; with \"ASYNC\", " + |
There was a problem hiding this comment.
This doc is intended for end-users, and I think rmm::mr::arena_memory_resource is too much detail here. I suggest simplifying this to say it uses the new CUDA stream-ordered memory allocator in CUDA 11.2+.
Signed-off-by: Rong Ou <rong.ou@gmail.com>
abellina
approved these changes
Sep 10, 2021
jlowe
approved these changes
Sep 10, 2021
|
build |
|
build |
1 similar comment
|
build |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
This only adds an option to set the gpu memory pool to use RMM
cuda_async_memory_resource. Additional work needs to be done to support UCX (e.g. openucx/ucx#7204) and possibly GDS.Depends on rapidsai/cudf#9201.
Signed-off-by: Rong Ou rong.ou@gmail.com