0

Reduce latency metric looks incorrect for binding pull requests in Milvus

While monitoring Milvus performance metrics, I noticed something unexpected related to binding pull requests.

Specifically, the reduce latency metric appears to be incorrect or misleading when the request type is a binding pull request. The actual request behavior does not show obvious performance issues, but the reported reduce latency value does not seem to reflect what is happening in practice.

So far, I’ve observed that:

  • The issue only occurs for binding pull requests

  • Other request types report reasonable latency metrics

  • There are no related errors or failed requests in the logs

This makes me wonder whether the reduce latency metric for binding pull requests is calculated differently or if this is a known issue in Milvus.

Is the reduce latency metric expected to behave differently for binding pull requests, or could this be an issue with how the metric is collected or reported?

Improve this question
Related questions
959
Why does C++ code for testing the Collatz conjecture run faster than hand-written assembly?
522
Why does GCC generate 15-20% faster code if I optimize for size instead of speed?
413
Why does the order of the loops affect performance when iterating over a 2D array?
Load 5 more related questions Show fewer related questions

0

Reset to default

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.