scottk/kubernetes
1
0
Fork 0
mirror of https://github.com/optim-enterprises-bv/kubernetes.git synced 2025-11-04 04:08:16 +00:00
Code Issues Packages Projects Releases Wiki Activity

doc(scheduler_perf): enrich the documentation

Browse Source
This commit is contained in:
Kensei Nakada
2024-01-14 07:57:23 +00:00
parent 12fc215656
commit f29d6970c9
1 changed files with 10 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
test/integration/scheduler_perf/README.md
View File

@@ -1,8 +1,11 @@
Scheduler Performance Test # Scheduler Performance Test
======
This package contains the scheduler performance tests, often called scheduler_perf.
We use it for benchmarking the scheduler with in-tree plugins, which is visible at [perf-dash](https://perf-dash.k8s.io/#/?jobname=scheduler-perf-benchmark&metriccategoryname=Scheduler&metricname=BenchmarkPerfResults&Metric=SchedulingThroughput&Name=SchedulingBasic%2F5000Nodes%2Fnamespace-2&extension_point=not%20applicable&result=not%20applicable).
Also you can use it outside the Kubernetes repository with out-of-tree plugins by making use of `RunBenchmarkPerfScheduling`.
## Motivation
Motivation
------
We already have a performance testing system -- Kubemark. However, Kubemark requires setting up and bootstrapping a whole cluster, which takes a lot of time. We already have a performance testing system -- Kubemark. However, Kubemark requires setting up and bootstrapping a whole cluster, which takes a lot of time.
We want to have a standard way to reproduce scheduling latency metrics result and benchmark scheduler as simple and fast as possible. We have the following goals: We want to have a standard way to reproduce scheduling latency metrics result and benchmark scheduler as simple and fast as possible. We have the following goals:
@@ -24,11 +27,9 @@ Currently the test suite has the following:
- make use of `go test -bench` and report nanosecond/op. - make use of `go test -bench` and report nanosecond/op.
- schedule b.N pods when the cluster has N nodes and P scheduled pods. Since it takes relatively long time to finish one round, b.N is small: 10 - 100. - schedule b.N pods when the cluster has N nodes and P scheduled pods. Since it takes relatively long time to finish one round, b.N is small: 10 - 100.
## How To Run
How To Run ### Benchmark tests
------
## Benchmark tests
```shell ```shell
# In Kubernetes root path # In Kubernetes root path
@@ -101,7 +102,7 @@ performance.
During interactive debugging sessions it is possible to enable per-test output During interactive debugging sessions it is possible to enable per-test output
via -use-testing-log. via -use-testing-log.
## Integration tests ### Integration tests
To run integration tests, use: To run integration tests, use:
``` ```