Calling cuda() consumes all the RAM memory

AastaLLL · July 27, 2021, 5:03am

Hi,

We test this on our Nano with JetPack 4.5.1 and PyTorch v1.8.0 installed from this topic.

RAM 1052/3964MB (lfb 91x4MB) ...
RAM 1052/3964MB (lfb 91x4MB) ...
RAM 1052/3964MB (lfb 91x4MB) ...
RAM 1130/3964MB (lfb 91x4MB) ...
RAM 1240/3964MB (lfb 91x4MB) ...
RAM 1376/3964MB (lfb 91x4MB) ...
RAM 1513/3964MB (lfb 91x4MB) ...
RAM 1665/3964MB (lfb 88x4MB) ...
RAM 1821/3964MB (lfb 80x4MB) ...
RAM 1999/3964MB (lfb 74x4MB) ...
RAM 2146/3964MB (lfb 36x4MB) ...
RAM 2315/3964MB (lfb 55x1MB) ...
RAM 2386/3964MB (lfb 15x1MB) ...
RAM 2386/3964MB (lfb 15x1MB) ...
RAM 2386/3964MB (lfb 15x1MB) ...

The memory increases around 1GiB in our experiment.
Is this similar to your observation?

Usually, the underlying library is loaded when being used.
To create a GPU buffer may trigger some requirements of the CUDA-related library, ex cuDNN.

Thanks.

Topic		Replies	Views
Pytorch in cuda mode use huge memory Jetson Nano pytorch	3	1388	September 5, 2021
JetPack 4.6.1 (L4T R32.7.1): PyTorch allocates all the memory + swap! Jetson Nano cuda , pytorch	5	626	December 19, 2023
Jetson nano sometimes extremely slow with GPU Jetson Nano cuda , pytorch	7	1351	November 3, 2023
GPU out of memory when the total ram usage is 2.8G Jetson TX2	28	18965	October 18, 2021
Official Tensorflow uses all RAM when running on GPU Jetson Nano	7	2356	October 18, 2021
Resource exhausted: OOM when allocating tensor with shape[256] Jetson Nano tensorrt , tensorflow	6	5443	October 18, 2021
Memory for GPU so small? Jetson TX2	14	4770	October 18, 2021
Question about the memory usage of jeston nanoo Jetson Orin Nano cuda , jetson-inference , python	3	136	October 17, 2024
General Question about Jetsons GPU/CPU Shared Memory Usage Jetson TX2	35	7913	October 18, 2021
slow to run tensorflow Resnet - how do I increase RAM available to GPU Jetson Nano	6	3868	October 18, 2021

Calling cuda() consumes all the RAM memory

Related topics