4paradigm/k8s-vgpu-scheduler

Issues

• vgpu-scheduler container kube-scheduler error

  #43 opened 3 months ago by G-Kolls
  2

• 消费级显卡支持么

  #42 opened 4 months ago by 13567436138
  0

• 支持新版资源调度么，比如resourceClaim方式

  #41 opened 4 months ago by 13567436138
  0

• 如何在Prometheus里监控gpu的使用情况

  #40 opened 6 months ago by efeng-blue
  0

• vgpu-device-plugin CreateContainerError

  #37 opened 6 months ago by thinkeng
  0

• 可以支持一下 kubeedge 么

  #38 opened 6 months ago by thinkeng
  0

• Failed to initialize NVML: could not load NVML library.

  #36 opened 8 months ago by zbjjyy
  1

• parameter devicePlugin.deviceSplitCount does not work

  #35 opened 8 months ago by 2232729885
  3

• 请问该调度器的实现方式是Extender吗？是否可以改为Scheduling Framework？

  #34 opened 9 months ago by CaptainLeezz
  0

• vgpu repo had not found

  #32 opened 9 months ago by swimtobird
  0

• undefined symbol:_dl_sym,version GLIBC_PRIVATE

  #30 opened a year ago by larry-lu-lu
  1

• failed calling webhook "vgpu.4pd.io"

  #29 opened 2 years ago by Tweakzx
  0

• Vgpu的限制问题

  #28 opened 2 years ago by changge27
  1

• 请问下 libvgpu.so 的代码 可以开源么?

  #26 opened 2 years ago by zhengborong
  0

• run nvidia-smi err in pod

  #25 opened 2 years ago by chenyangxueHDU
  0

• core dump when request 2 or more gpus with Tesla T4

  #24 opened 2 years ago by ryan1051
  1

• Who's using vGPU K8s Device Plugin / 您在使用vGPU K8s Device Plugin吗 ?

  #5 opened 3 years ago by xuman2019
  2

• Handle_remap not found handle

  #22 opened 2 years ago by RexQian
  0

• 能否支持像cpu那样按微核分配资源，比如我把一块划分为100份，我一个pod可以指定任意份数，比如“10m”或“20m”

  #21 opened 3 years ago by fu7100
  0

• how to install in openshift4 ?

  #20 opened 3 years ago by fu7100
  0

• can't find function nvmlDeviceGetComputeRunningProcesses_v2 in libnvidia-ml.so.1

  #4 opened 3 years ago by haijohn
  3

• Error: failed to create FS watcher: no such file or directory

  #6 opened 3 years ago by qifengz
  2

• 我使用的是v0.9.0.0这个版本，build之后，部署为daemon服务到 GPU节点, 报device-split-count等几个参数未定义，去掉这几个参数后，POD可正常在GPU节点running；但看日志找到不到NVML,GPU节点是P100,求联系求指导

  #9 opened 3 years ago by AlexPei
  1

• 切分10份，但是VGPU显存无变化 NVIDIA A100

  #19 opened 3 years ago by Dripman
  4

• Is there a way to monitor vGPU with DCGM?

  #18 opened 3 years ago by rjanovski
  1

• 分配2张vgpu却只能看到1张

  #17 opened 3 years ago by xwhuang0923
  5

• Why the server could not find the requested resource？

  #15 opened 3 years ago by airaer1998
  1

• 两张GPU，只识别了一张卡

  #16 opened 3 years ago by absolutelyZero
  1

• Segmentation fault (core dumped)

  #14 opened 3 years ago by bingMillion
  0

• can use for 2080Ti/1080Ti ？

  #13 opened 3 years ago by sunhao12121
  2

• 启用vGPU之后pod内执行nvidia-smi报错Segmentation fault (core dumped)

  #11 opened 3 years ago by detongz
  3

• 显存隔离

  #12 opened 3 years ago by alexk1028
  4

• 关于使用的疑惑

  #10 opened 3 years ago by alexk1028
  2

• commited image can not run in another node.

  #8 opened 3 years ago by haijohn
  3

• 显存显示问题

  #7 opened 3 years ago by qifengz
  8

• 能否增加选择指定GPU切分

  #2 opened 3 years ago by GuoYingLong
  1

• 切分功能不起作用，请求帮助？

  #3 opened 3 years ago by absolutelyZero
  0