Improve benchmark tput by moving prompt preparation outside of loop #54
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
Moved the prompt making call to randomly_sample_sonnet_lines_prompt outside of load request send loop so that the send loop can generate load to the server faster. Otherwise there's an artificial delay due to making the next prompt which slows down the benchmark throughput/sec.
Also changed tokenizer instantiation to just once outside the prompt generation loop to speed up the overall test.
After this change I've seen up to 2x improvement in server achieved throughput in some small workloads. This change will allow better measurement of true server throughput.