1.0 What's New
1. GPU Sharing
Volcano now supports gpu sharing between different pods (#852, @tizhou86, @hzxuzhonghu).
2. Preempt and reclaim enhancement
Volcano is now able to support preempt for batch job (#738, @carmark).
3. Dynamic scale up and down
Volcano job now supports dynamically scale up and down (#787, @hzxuzhonghu).
4. Support integrate with flink operator
Users are now able to run flink job with volcano. Follow the instructions here to make use of the feature. @hzxuzhonghu).
5. Support DAG job with argo
Users are now able to run DAG job with volcano. Follow the instructions here to make use of the feature. @alcorf-mizar).
Other Notable Changes
- Update go version to 1.14 (#886, @hzxuzhonghu)
- Bump to k8s 1.18 to keep up with kubernetes (#855, @hzxuzhonghu)
- Add mindspore example (#845, @lyd911)
- Add golangCi-lint check (#799, #821, #824, #825, #827, #829, #830, #833, #835, #841. @masihtehrani @daixiang0 @Thor-wl)
- Allow specifying admission webhook port (#832, @hzxuzhonghu)
- Add support for most requested priority (#831, @daixiang0)
- Set pod DNSPolicy to ClusterFirstWithHostNet when hostnetwork set (#779, @hzxuzhonghu)
- Add bestNodeFn for plugins to select best node of its own (#790, @jiangkaihua)
- Add e2e cases for drf (#905, @Thor-wl)
- Add e2e cases for reclaim (#898 #906, @alcorj-mizar)
- Add e2e cases for preempt (#892, @Thor-wl )
- Add e2e for queue (#872, @Thor-wl)
- Add test code for queue controller (#858, @alcorj-mizar)
Bug Fixes
- Fix panic in controller (#903, @Thor-wl)
- Fix panic in allocate (#843, @k82cn)
- Fix job phase transition time set (#789, @hzxuzhonghu)
- Fix crd to support job patch ops (#786, @hzxuzhonghu)