Support top level response caching for ensemble models #338
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
Tabrizian
reviewed
Apr 4, 2024
src/ensemble_scheduler/ensemble_scheduler/ensemble_scheduler.cc
Outdated
Show resolved
Hide resolved
rmccorm4
reviewed
Apr 4, 2024
rmccorm4
reviewed
Apr 4, 2024
This was referenced Apr 5, 2024
lkomali
changed the title
rmccorm4
reviewed
Apr 8, 2024
|
FYI don't merge this until the testing PR is also ready to merge and pipelines look good |
lkomali
added a commit
that referenced
this pull request
May 9, 2024
mc-nv
added a commit
that referenced
this pull request
May 10, 2024
mc-nv
pushed a commit
that referenced
this pull request
May 10, 2024
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Outdated suggestions cannot be applied.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.
ref slack thread: https://nvidia.slack.com/archives/CAZKCU4UV/p1677717244222069
Currently caching at the top-level request sent to ensemble scheduler is not supported.
Implemented caching top level requests for ensemble models.
In case of cache hit, the response is sent without executing composing models.
In case of cache miss, the ensemble pipeline runs as is.
Changed Logic for computing Cache Miss Latency: Cache Lookup time + Ensemble Pipeline time + Cache Insertion time
Moved similar logic form ensemble_scheduler.cc and dynamic_batch_scheduler.cc to scheduler_utils.cc to reduce code redundancy.