v2.8.0
๐ Major featuresโ
- HAMi DRA is now supported, for details, please visit: https://github.com/Project-HAMi/HAMi-DRA
- Enable leader select among multiple schedulers (#1553)
- Support CDI mode on NVIDIA devices (#1552)
- Optimize HAMi webUI, please visit https://github.com/Project-HAMi/HAMi-WebUI
- Sync with k8s-device-plugin from nvidia v0.18.0 (#1541)
- Add hami_build_info metrics and version print (#1581)
- Watch and hot reload the updated certificate (#1573)
๐ Major bug fixesโ
- Update HAMi-core to fix vllm-related issues: #1381 #1461 by (@archlitchi) in #1478
- Fix: Calculation error for quotas by (@luohua13) in #1400
- Fix: vXPU feature may not working properly on P800 node (#1569)
- Fix scheduler allocate incorrect mig instance (#1518)
๐ What's Changedโ
๐จ Other Changesโ
- Mock-device-plugin is now ready to use, please visit: https://github.com/Project-HAMi/mock-device-plugin
- Ascend device plugin is now supporting vNPU feature for both HAMi and volcano, please visit: https://github.com/Project-HAMi/ascend-device-plugin
- Refine Node Register logic (#1499)
- Update go version to v1.25.5
- Fix release CI by (@archlitchi) in #1373
- Fix: failed clusterrolebinding when change release name or chart name by (@FouoF) in #1380
- fix: e2e ginkgo version mismatch by (@FouoF) in #1391
- fix: check pod nil in
ReleaseNodeLockby (@DSFans2014) in #1372 - fix: upgrade nvidia-mig-parted to v0.12.2 to solve security issues by (@Shouren) in #1388
- fix: scheduler flaky test by (@FouoF) in #1402
- Fix: After removing the device plugin from the gpu node, it can stillโฆ by (@luohua13) in #1456
- Fix concurrent map iteration and map write fatal error. by (@litaixun) in #1452
- fix: fix typos by (@DSFans2014) in #1434
- Fix CI error of the PR #1470, #1326, #1033 by (@archlitchi) in #1473
- Fix concurrent map read write fatal error. by (@litaixun) in #1476
- add podInfos in DeviceUsage to enhance scheduling decision by (@Kyrie336) in #1362
- Update device-numa acquisition logic by (@archlitchi) in #1403
- Improved support for iluvatar GPUs by (@qiangwei1983) in #1399
- Improve: Replace
StrategicMergePatchTypebyMergePatchTypeby (@luohua13) in #1431 - optimize schedule failure event by (@Kyrie336) in #1444
Committers: Contributorsโ
- archlitchi (@archlitchi)
- FouoF (@FouoF)
- DSFans2014 (@DSFans2014)
- Shouren (@Shouren)
- luohua13 (@luohua13)
- litaixun (@litaixun)
- Kyrie336 (@Kyrie336)
- qiangwei1983 (@qiangwei1983)
Full Changelog: https://github.com/Project-HAMi/HAMi/compare/v2.7.1...v2.8.0