CPU Pinning and Numa Tuning ConnectX_5/ConnectX_6 notebook not working

[Rasman Mubtasim Swargo]

I’m trying to run the CPU Pinning and NUMA Tuning (ConnectX-5/ConnectX-6) notebook with a few modifications because I need an NVMe drive and a ConnectX-6 NIC for my experiments. I have always used this notebook to create slices, and it used to work.

Since last week, I’ve noticed that I can’t ping or transfer files from node1 to node2 created with the same notebook. I attached the notebook here. Even the iperf3 cell won’t run/connect.

Could you help me understand what I might be doing wrong or what changed?

[Rasman Mubtasim Swargo]

Sorry, I could not upload the notebook as the file type is not supported.

Here is the github link of the notebook: https://github.com/swargo98/LLM-based-Data-Movement-Optimizer/blob/main/iperf3_optimized_w_error.ipynb

[Komal Thareja]

Hi Rasman,

I was able to run iperf3 optimized notebook without issues. I am unable to access your notebook. It says Page Not Found.

Could you please share your slice ID?

Thanks,

Komal

[Rasman Mubtasim Swargo]

Komal Thareja wrote:

Hi Rasman,

I was able to run iperf3 optimized notebook without issues. I am unable to access your notebook. It says Page Not Found.

Could you please share your slice ID?

Thanks,

Komal

Hi Komal,
Here is the slice ID: f05dedc0-468f-406e-a566-8041d507ad60

I am trying to transfer some files from MASS to TACC but it is failing to find any route.

 

Here is the notebook that I used: https://github.com/swargo98/LLM-based-Data-Movement-Optimizer/blob/main/iperf3_optimized_w_error.ipynb

[Rasman Mubtasim Swargo]

Komal Thareja wrote:

Hi Rasman,

I was able to run iperf3 optimized notebook without issues. I am unable to access your notebook. It says Page Not Found.

Could you please share your slice ID?

Thanks,

Komal

I have tried again, this time without any modifications to the original notebook, but I still got the same error.

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Source:  Node-MASS to Dest: Node-TACC
iperf3: error - unable to connect to server: No route to host
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Source:  Node-TACC to Dest: Node-MASS
iperf3: error - unable to connect to server: No route to host
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

 

Slice ID: 919b590f-87aa-41e8-bb22-58acf4c79d4c

[Komal Thareja]

Hi Rasman,

The dataplane was down at MASS. The link has been restored, could you please try running the iPerf3 cells again from the notebook. Apologies for the inconvenience.

Thanks,

Komal

[Rasman Mubtasim Swargo]

It’s working now.

Thanks,

Swargo

[Author]

Posts