Load test recipes with cluster loader 2.#267
Load test recipes with cluster loader 2.#267SHRUTI6991 wants to merge 8 commits intokubernetes-sigs:mainfrom
Conversation
…nd calculate it's latency.
✅ Deploy Preview for agent-sandbox ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
k8s-ci-robot
added
cncf-cla: yes
|
Hi @SHRUTI6991. 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. DetailsInstructions 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. |
k8s-ci-robot
added
the
size/XL
janetkuo
added
ok-to-test
| - type: Running | ||
| labelSelector: "agents.x-k8s.io/pool" | ||
|
|
||
| # --- 2. DRAIN WARMPOOL (Excluded from Cold Measurement) --- |
There was a problem hiding this comment.
I understand the intent is to establish a baseline for Cold Start latency after exhausting the warm pool. But just curious shouldn't we also measure the drain phase (Step 2)? Because afaiu, the test tells us 'How bad is it when the pool is empty?' (Cold Start). But it doesn't tell us 'How good is it when the pool is full?' (Warm Acquisition).
| latency-type: {{.LatencyType}} | ||
| spec: | ||
| restartPolicy: Never | ||
| runtimeClassName: gvisor |
There was a problem hiding this comment.
Can we template this to {{.RuntimeClass}} ? This would allow us to run these load tests against local clusters (e.g., Kind with runc) or accept other runtimeClassNames in general. It could also help for CI verification as well.
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: aditya-shantanu, SHRUTI6991, vicentefb The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
5 participants
This PR creates 4 custom test recipes that performs 4 different tests. These test can be used to run different scenarios with different cloud providers.
throughput-test- This test allows varying the creation rate QPS and checking how QPS variation affects Pod scheduling and ready rate.medium-scale-concurrent-load-test- This test allows the simultaneous creation and deletion of sandboxes to determine the latency during varying sustained loads.high-volume-test- This test linearly scales up sandboxes at a constant QPS to check how increasing the sandboxes affect agent controller and the cluster.warmpool-burst-test- This test calculates the pod acquisition latency for cold start. Currently we don't emit any warm pool acquisition metrics.