Memory consumption is not consistent with Multi-GPU

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) Titan
• DeepStream Version 5
• TensorRT Version 7.0.0.11
**• NVIDIA GPU Driver Version (valid for GPU only)**440.59
• Issue Type( questions, new requirements, bugs) Questions

I have 4 gpus. I am running the same deep-stream configuration file with the same models on all 4 at the same time. Why is the memory consumption is not consistent on all 4 gpus?. They are different.

Also, memory consumption of GPU0 increases when I run other gpus. why? I thought each gpu is independent.

image
image984×527 247 KB

3

Hi
Please specify details how to recreate your senario?
Did you run 4 processes using same configuration? if so, can you provide the configurations you used?

4

I have similar issue with you. What’s you configuration for [sink] ?