关于GCU、沐曦GPGPU、MLU、0卡V100资源4月7日恢复上架的公告>>> 关于共建具身智能开源数据集的倡议>>> 关于云脑任务中统一路径访问方式的公告>>> 关于将启智集群GPU资源迁移至智算集群的公告>>>

New Issue

#4081 智算GCU集群训练任务支持查看资源利用率和运行简况

Closed

created 1 year ago by avadesian · 1 comments

tanglj added the

enhancement

label 11 months ago

tanglj added this to the V20230531 milestone 11 months ago

liuzx was assigned by tanglj 11 months ago

zhoupzh was assigned by tanglj 11 months ago

chenshihai was assigned by tanglj 11 months ago

zhoupzh was unassigned by tanglj 11 months ago

chenyifan01 was assigned by tanglj 11 months ago

liuzx was unassigned by tanglj 11 months ago

wangj modified the milestone from V20230531 to V20230628 10 months ago

wangj modified the milestone from V20230628 to V20230718 10 months ago

chenshihai added this to the ai_task_refactor_v3_train branch 9 months ago

chenshihai added the

test

label 9 months ago

wangj was assigned by chenshihai 9 months ago

通过测试。
gcu训练的资源占用情况曲线，图例仍然是gpuMemUsage、gpuUtil。已确认，是章鱼接口返回暂时没有gcu的；
gcu训练的任务运行简况，Scheduled、Pulled、 Created、Started、Exit这几个阶段，有时候会少显示几个。已向章鱼反馈。

wangj closed this issue 9 months ago

test

V20230718

2 Participants

Notifications

Due Date

No due date set.

Dependencies

This issue currently doesn't have any dependencies.