[Bug] Network traffic monitor

[zheng-ningxin]

Organization Name: Microsoft

Short summary about the issue/question:
I submit a job that occupies a whole GPU node. Basically, this job trains a classification model on imagenet, and the imagenet dataset is stored on the corresponding nfs storage. I want to monitor the network bandwidth consumed by this job and find that the consumed network bandwidth is zero.

Then I change to the node-view and find that the consumed network bandwidth is not zero this time(the job started at 11.05). So I think the network bandwidth monitoring script may have some bugs.

Brief what process you are following:

How to reproduce it:

OpenPAI Environment:

• OpenPAI version:
• Cloud provider or hardware configuration:
• OS (e.g. from /etc/os-release):
• Kernel (e.g. uname -a):
• Hardware (e.g. core number, memory size, storage size, GPU type etc.):
• Others:

Anything else we need to know: