Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

wip: tracing/metrics #122

Closed
wants to merge 1 commit into from
Closed

wip: tracing/metrics #122

wants to merge 1 commit into from

Conversation

kylebarron
Copy link
Member

For #105

Well, this is way more complex than I originally thought.

I'm interested in having basic request metrics for an object store: number of requests, their type (get/list/put/delete), timing, and number of bytes downloaded.

At first I was thinking of doing this with a simple Mutex<Vec<RequestStatistics>>, but realized it would lead to horrible lock contention when each request wants to append to the vec.

So I've been looking at the tracing crate and its Instrument, which looks like it might perform better for async code.

However it looks like that's considerably more complex than I want to get into right now.

In theory, a mature implementation of this would allow the Python user to choose where the traces actually go. Like you could set this up to use ObjectStore but send tracing information to, say, CloudWatch

I'll put this stub up just to record my thoughts.

@kylebarron
Copy link
Member Author

Closing in favor of looking at the influxdb implementation: https://github.com/influxdata/influxdb3_core/blob/main/object_store_metrics/src/lib.rs

@kylebarron kylebarron closed this Jan 8, 2025
@kylebarron kylebarron mentioned this pull request Jan 15, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant