tkestack/vcuda-controller

Issues

• 进入到 pod 里面 执行 nvidia-smi ， 显示的显存 和 node 节点是一致， pod 里面应该显示 给pod 分配的资源吧

  #26 opened 3 years ago by khw934
  2

• get "invalid device context" or "segmentation fault" problems in aarch64 machine

  #35 opened 2 years ago by DennisYoung96
  4

• torch:1.12.0+cu113 驱动:530.41.03，gpu manager调度成功之后，使用cuda报错

  #34 opened 2 years ago by Justin-ZL
  3

• support cuda 11.7

  #36 opened a year ago by AdrianMachao
  5

• Failed to compile

  #39 opened a year ago by 283713406
  1

• 导出tensorrt模型报错

  #38 opened a year ago by vicmeng
  0

• Why are these symbols marked as "deprecated"?

  #37 opened a year ago by amet13x
  4

• P4 GPU can not get enough gpu utilization

  #12 opened 4 years ago by zwpaper
  15

• how to view logs output by vcuda-controller

  #27 opened 2 years ago by hzliangbin
  2

• CUBLAS_STATUS_ALLOC_FAILED (cuda11.4.4，cuBLAS 11.6.5.2)

  #33 opened 2 years ago by work-chausat
  3

• need update for cuda11.4？

  #17 opened 2 years ago by difenbei
  6

• Doubts about cuArray3DCreate_helper calculating memory usage

  #18 opened 2 years ago by rv64m
  3

• some issues on cuda-11.4

  #20 opened 2 years ago by M201972777
  3

• Problems caused by launching multiple pods at the same time

  #25 opened 3 years ago by rv64m
  6

• some questions about initialization

  #24 opened 3 years ago by matinjugou
  0

• why is nvmlDeviceSetComputeMode disabled when vcuda is enabled?

  #23 opened 3 years ago by matinjugou
  0

• Failed to parse cgroups path

  #19 opened 3 years ago by rainfd
  1

• untraceable GPU memory allocation

  #5 opened 5 years ago by zw0610
  13

• [Question] Why the `blocks` not actually used in `rate_limiter` ?

  #14 opened 4 years ago by lining2020x
  6

• How to use the vcuda by docker run

  #16 opened 4 years ago by MC17
  1

• A question about the function nvmlDeviceGetProcessUtilization

  #15 opened 4 years ago by hejinsong
  0

• Usage is not clear

  #6 opened 5 years ago by raz-bn
  24

• Which GPU are you using ?

  #13 opened 4 years ago by xiazhiyi99
  2

• can't find function libcuda.so.440.100 in cuEGLInit

  #10 opened 4 years ago by nlnjnj
  1

• nvprof hangs in vcuda

  #8 opened 4 years ago by openqt
  1

• The program based on TensorRT is slowing down with vcuda

  #9 opened 4 years ago by rotorliu
  1

• ECC status shows ERR！in vcuda container

  #7 opened 4 years ago by Eric-918
  1

• nvmlDeviceGetProcessUtilization disappers from 440

  #3 opened 5 years ago by zw0610
  1

• Add build guide and update the image name

  #1 opened 5 years ago by ruyingzhe
  1