The major target of this release to make Volcano more stable for product.
Improve test coverage. In v1.0, more test cases will be added to improve Volcano stability for product.
Preemption and Reclaim are two import features for resource sharing; there're two actions for now, but unstable. In v1.0, those two features are going to be enhanced for elastic workload, e.g. stream job, bigdata batch job.
GPU Share (#624)
A better performance has its cost, including GPU; and there are several scenarios that a Pod can not consume one GPU, e.g. inference workload, dev environment. One of solutions is to support GPU share, including related enhancement to both scheduler and kubelet.
Flink is a widely used for Stateful Computations over Data Streams, but flink on kubernetes has some gaps now.
Investigate to cooperate with argo to support job dependencies.
MindSpore is a deep learning training and inference framework, support running MindSpore training with volcano job.
- Description: Support reserve specified resource for queue without restart Volcano.
- Priority: High
- Issue: volcano-sh/volcano#1101
- Owner: @hudson741@Thor-wl
- Description: Schedule jobs of same priority but from different queue accord to create time.
- Priority: High
- Issue: volcano-sh/volcano#1098
- Owner: @alcorj-mizar
- Description: Support configure actions and plugins for different queues.
- Priority: High
- Issue: volcano-sh/volcano#1035
- Owner: @sresthas
- Description: Support Hierarchy Queue algorithm.
- Priority: High
- Issue: volcano-sh/volcano#1033
- Owner: @My-pleasure
- Description: Support update vcjob priorityClassName update when job has not been scheduled.
- Priority: Middle
- Issue: volcano-sh/volcano#1097
- Owner: @merryzhou
- Description: Provide more status detail for CRD status when use CLI such job fail reason.
- Priority: Middle
- Issue: volcano-sh/volcano#1094
- Owner:@mikechengwei
- Description: Support MinAvailable for task
- Priority: Middle
- Issue: volcano-sh/volcano#988
- Owner: @shinytang6
- Description: Support task topology scheduling
- Priority: High
- Issue: volcano-sh/volcano#1349
- Description: Support multiple scheduler by admission controller.
- Priority: High
- Issue: volcano-sh/volcano#1322
- Owner: @Thor-wl @zen-xu
- Description: Improve the UT/E2E test coverage and add the stress test to improve stability.
- Priority: High
- Issue: volcano-sh/volcano#1284
- Owner: @rudeigerc
- Description: Support container using multiples GPU as well as part of GPU card.
- Priority: High
- Issue: volcano-sh/devices#12
- Owner: @peiniliu
- Description: Support reserve specified resource for queue without restart Volcano.
- Priority: Medium
- Issue: volcano-sh/volcano#1101
- Owner: @Thor-wl @hudson741
- Description: Support Hierarchy Queue algorithm.
- Issue: volcano-sh/volcano#1033
- Description: Add hot update for Volcano components arguments.
- Issue: volcano-sh/volcano#1326
- Description: Support high accurate resource calculation.
- Issue: volcano-sh/volcano#1196
- Description: Add backfill functionality to improve the resource utilization.
- Description: Combine the Autoscaler and scheduler to improve the scaling efficiency.
- Description: Enhance monitor to add more metrics for jobs, queues, etc.
- Description: Add the stress test.