-
Notifications
You must be signed in to change notification settings - Fork 85
Introduce SaturationDetector component #808
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
Introduce SaturationDetector component #808
Conversation
✅ Deploy Preview for gateway-api-inference-extension ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
Hi @LukeAVanDrie. Thanks for your PR. I'm waiting for a kubernetes-sigs member to verify that this patch is reasonable to test. If it is, they should reply with Once the patch is verified, the new status will be reflected by the I understand the commands that are listed here. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
f10f62f
to
37e73c4
Compare
/ok-to-test |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks pretty good! Just some smaller comments, thanks!
be326b9
to
5217359
Compare
/retest |
/approve Will wait for tests/ response to comments before LGTM. Thanks Luke! |
Test failure is from an unrelated flake. I sent out a fix in #824 |
/lgtm |
This commit adds a new `SaturationDetector` component responsible for determining if backend model servers are saturated. It bases its decision on observed metrics like queue depth and KV cache utilization, using configurable thresholds. The detector is designed to be a self-contained unit that can be leveraged by other components for admission control and capacity assessment. This is the first step in a larger refactoring to externalize and centralize saturation detection logic.
5217359
to
33342b6
Compare
I cannot keep up with @nirrozenbaum :) Rebased on to the MetricsState change which should fix the failing tests. |
/lgtm |
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: kfswain, LukeAVanDrie The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
This commit adds a new `SaturationDetector` component responsible for determining if backend model servers are saturated. It bases its decision on observed metrics like queue depth and KV cache utilization, using configurable thresholds. The detector is designed to be a self-contained unit that can be leveraged by other components for admission control and capacity assessment. This is the first step in a larger refactoring to externalize and centralize saturation detection logic.
This commit adds a new `SaturationDetector` component responsible for determining if backend model servers are saturated. It bases its decision on observed metrics like queue depth and KV cache utilization, using configurable thresholds. The detector is designed to be a self-contained unit that can be leveraged by other components for admission control and capacity assessment. This is the first step in a larger refactoring to externalize and centralize saturation detection logic.
This commit adds a new
SaturationDetector
component responsible for determining if backend model servers are saturated. It bases its decision on observed metrics like queue depth and KV cache utilization, using configurable thresholds.To be submitted before #805 (split this change out for easier review).
Related to: #674
It's initial implementation uses identical thresholds to the
HasCapacity
filter embedded in the default Scheduler filter chain for sheddable requests.The detector is designed to be a self-contained unit that can be leveraged by other components for admission control and capacity assessment. This is the first step in a larger refactoring to externalize and centralize saturation detection logic.