Kaiyu Shi

Ter, The most time is not spent on parsing CLI options actually, which only requires several us at most. > On Mar 18, 2021, at 10:42 PM, Ter ***@***.***> wrote:...

What's the output of `nvidia-smi`

Unfortunately this tool (as well as any other GPU management tools) depends on the Nvidia driver and toolkit, you should at least install the Nvidia-driver to get things work. >...

@radhikasethi2011 Your problem seems like windows compatibility issue from PyNVML. I don't have any windows GPU server by my side, so I'm afraid I cannot fix it my self. But...

Great. Is it possible to add a test case for this bug fix?

When the json output is stable, maybe we can detach the data collecting module and display module. Then we can request GPU usages from different remote data collecting nodes via...

I'd like to recommend an NVML binding for Python to collect GPU information rather than `nvidia-smi`. This library would significantly simplify the code base for information gathering. https://github.com/jonsafari/nvidia-ml-py

I'm trying to move the `nvidia-smi` API to `nvidia-ml-py` Update: you may like to review my PR #17

@gyscos I think fileds with dot can be naturally organized into sub-dict. e.g. ```json { 'utilization': { 'gpu': 40, 'memory': 30, } } ```

Prefer a list-like argument e.g. `--show power, name, temp`, which overwrites other displaying options. But it's still confusing with the existing `-cpu` arguments, the latter one is for extra information.