Azure / Azure/AzurePublicDataset

Missing prefix cache hit rate in Azure LLM inference trace 2023 and 2024.

Open
#53 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Jupyter Notebook
Stars
1.2k
Forks
185
PR merge metrics
No merged PRs in 30d

Description

In these two traces, there are only input and output lengths, but no prefix cache hit rate. Therefore, the data tested using these traces cannot truly reflect the inference performance and load conditions. Could you please supplement these traces with their corresponding prefix cache hit rates?

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.