What's the procedure for the GPU metrics reporting script setup?

0 votes
Can somebody help me with the set up of GPU metrics reporting script?
Nov 7 in GCP by anonymous
• 5,700 points
12 views

1 answer to this question.

0 votes

To Set up the GPU metrics reporting script:

On each of your VM instances, check that you meet the following requirements:

  1. Each VM instance must have GPUs attached.

  2. Each VM instance must have a GPU driver installed.

  3. Each VM instance must have the pip utility installed. On each of your VM instances, install the GPU metrics agent. 

  4. To install the metrics agent, complete the following steps: 

  • Download the GPU metrics reporting scripts.

git clone https://github.com/GoogleCloudPlatform/tensorflow-inference-tensorrt5-t4-gpu.git 

  • Switch to the metrics_reporting folder.

cd tensorflow-inference-tensorrt5-t4-gpu/metrics_reporting
  • Set up the installation environment for the metrics agent.

pip install -r ./requirements.txt
  • Move the metric reporting script to your root directory.

sudo cp report_gpu_metrics.py /root/
  • Enable the GPU metrics agent.

cat <<-EOH > /lib/systemd/system/gpu_utilization_agent.service
[Unit]
Description=GPU Utilization Metric Agent
[Service]
Type=simple
PIDFile=/run/gpu_agent.pid
ExecStart=/bin/bash --login -c '/usr/bin/python /root/report_gpu_metrics.py'
User=root
Group=root
WorkingDirectory=/
Restart=always
[Install]
WantedBy=multi-user.target
EOH
  • Reload the system daemon.

systemctl daemon-reload
  • Enable the gpu monitoring service.

systemctl --no-reload --now enable /lib/systemd/system/gpu_utilization_agent.service
git clone https://github.com/GoogleCloudPlatform/tensorflow-inference-tensorrt5-t4-gpu.git
systemctl --no-reload --now enable /lib/systemd/system/gpu_utilization_agent.service
answered Nov 7 by anonymous
• 38,820 points

Related Questions In GCP

0 votes
1 answer
0 votes
1 answer
0 votes
1 answer

How do I add the access config for my Linux VM instance?

If your instance does not have an ...READ MORE

answered Sep 26 in GCP by anonymous
• 5,700 points
36 views
0 votes
1 answer
0 votes
1 answer

Creating a SQL Server instance using Google Compute engine.

Google Compute Engine provides public images preconfigured with ...READ MORE

answered Sep 23 in GCP by Sirajul
• 38,820 points
38 views
0 votes
1 answer

How do i install gcloud compute?

The gcloud compute command-line tool enables you to easily ...READ MORE

answered Sep 23 in GCP by Sirajul
• 38,820 points
48 views
0 votes
1 answer

where is drop down menu for the google cloud platform

It is important that you are signed ...READ MORE

answered Sep 20 in GCP by Sirajul
• 38,820 points
60 views