kubernetes/minikube

Generate performance metric for minikube start/stop/delete speed on different drivers

Open

#4,622 opened on 2019年6月27日

GitHub で見る
 (5 comments) (3 reactions) (0 assignees)Go (31,799 stars) (5,222 forks)batch import
area/performancehelp wantedkind/featurelifecycle/frozenpriority/backlog

説明

it would be a great thing to have a script to run on each release, that does the following:

  • measure how long minikube start, start again ,stop , delete takes on each different drivers such as, virtualbox, hyperkit, hyperv and none.

the script should generate the a dirver_performance.md file in the /docs automatically on each release and purpose a PR.

The script could run as an integration test. the same way that our e2e tests run.

コントリビューターガイド

技術スタック
go
領域
performancedevopstesting
Issue 種別
feature
難度新しい貢献者にとっての実装難度です。1 は非常に小さな変更、5 は専門的な作業です。
3
推定時間経験ある貢献者が調査、実装、テスト、pull request 準備にかけるおおよその時間範囲です。
over 1 week
活動状況Issue が今どれだけ取り組みやすいかを示します。新しい、活発、古い、ブロック中、またはメンテナー入力待ちなどです。
stale
明確さ期待される変更、受け入れ条件、次の手順がどれだけ明確かを示します。
mostly clear
前提条件
Go programmingminikube internalsknowledge of virtualization driverse2e testing experience
初心者向け度初回貢献者にどれだけ取り組みやすいかを 1-100 で推定したスコアです。
40
調査方針
The issue requests a script to measure minikube start/stop/delete speed on different drivers (virtualbox, hyperkit, hyperv, none) and auto generate a driver performance.md file in /docs on each release. To begin, examine existing e2e tests in the minikube repository (e.g., integration tests under test/integration/) to understand the test harness and how they invoke minikube commands. The script should be similar to existing tests but focus on timing operations. Look at the CI configuration (e.g., in .github/workflows) to see how tests are run on different drivers. The deliverable is a script that can be integrated into the release process, potentially as a new CI job. Note that this is a complex feature that requires understanding of multiple drivers and the release pipeline.