[GPU] Extend CuVSResourcesManager #137588
Merged
Conversation
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
ldematte
added
>non-issue
auto-backport
Collaborator
|
Pinging @elastic/es-search-relevance (Team:Search Relevance) |
elasticsearchmachine
added
Team:Search Relevance
ChrisHegarty
approved these changes
Nov 4, 2025
x-pack/plugin/gpu/src/main/java/org/elasticsearch/xpack/gpu/codec/TrackingGPUMemoryService.java
Show resolved
Hide resolved
x-pack/plugin/gpu/src/main/java/org/elasticsearch/xpack/gpu/codec/CuVSResourceManager.java
Show resolved
Hide resolved
x-pack/plugin/gpu/src/main/java/org/elasticsearch/xpack/gpu/codec/CuVSResourceManager.java
Show resolved
Hide resolved
Collaborator
💚 Backport successful
|
ldematte
added a commit
to ldematte/elasticsearch
that referenced
this pull request
Nov 5, 2025
CuVSResourcesManager has the purpose of controlling access to resources to ensure a correct level of parallelism (allowing more than 1 GPU thread, but having a reasonable upper bound) and controlling the amount of GPU memory needed to prevent CUDA out-of-memory errors.
This PR extends the memory control part by introducing different strategies for memory accounting ("real", based on API calls to the device, and "tracking", which remembers the amount of memory requested during acquisition) and different estimations based on the CAGRA graph build algorithm.
The former will allow us to use pooled memory (where the amount of available memory will be different from the free device memory), the latter to use the IVFPQ CAGRA graph build algorithm for larger datasets.
elasticsearchmachine
pushed a commit
that referenced
this pull request
Nov 5, 2025
CuVSResourcesManager has the purpose of controlling access to resources to ensure a correct level of parallelism (allowing more than 1 GPU thread, but having a reasonable upper bound) and controlling the amount of GPU memory needed to prevent CUDA out-of-memory errors.
This PR extends the memory control part by introducing different strategies for memory accounting ("real", based on API calls to the device, and "tracking", which remembers the amount of memory requested during acquisition) and different estimations based on the CAGRA graph build algorithm.
The former will allow us to use pooled memory (where the amount of available memory will be different from the free device memory), the latter to use the IVFPQ CAGRA graph build algorithm for larger datasets.
Kubik42
pushed a commit
to Kubik42/elasticsearch
that referenced
this pull request
Nov 10, 2025
CuVSResourcesManager has the purpose of controlling access to resources to ensure a correct level of parallelism (allowing more than 1 GPU thread, but having a reasonable upper bound) and controlling the amount of GPU memory needed to prevent CUDA out-of-memory errors.
This PR extends the memory control part by introducing different strategies for memory accounting ("real", based on API calls to the device, and "tracking", which remembers the amount of memory requested during acquisition) and different estimations based on the CAGRA graph build algorithm.
The former will allow us to use pooled memory (where the amount of available memory will be different from the free device memory), the latter to use the IVFPQ CAGRA graph build algorithm for larger datasets.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
CuVSResourcesManagerhas the purpose of controlling access to resources to ensure a correct level of parallelism (allowing more than 1 GPU thread, but having a reasonable upper bound) and controlling the amount of GPU memory needed to prevent CUDA out-of-memory errors.This PR extends the memory control part by introducing different strategies for memory accounting ("real", based on API calls to the device, and "tracking", which remembers the amount of memory requested during acquisition) and different estimations based on the CAGRA graph build algorithm.
The former will allow us to use pooled memory (where the amount of available memory will be different from the free device memory), the latter to use the IVFPQ CAGRA graph build algorithm for larger datasets.