See the tag entry for "multi-gpu".
Questions tagged [multiple-gpu]
43 questions
0
votes
0 answers
How to specify gpu device numbers when using accelerate's notebook_launcher?
When using Accelerate's notebook_launcher to kickoff a training job spawning across multiple GPUs, is there a way to specify which GPUs (i.e. CUDA_VISIBLE_DEVICES=“4,5,6,7”) to be used, in stead of starting with default cuda:0?
from accelerate…
alfred Shen
- 11
- 2
0
votes
0 answers
Possible to use tf.distribute.Strategy.mirroredstrategy on parts of the graph rather than entire train_step for GAN custom training script?
I'm writing CTGAN code and want it to train in a distributed way.
Therefore I'm using tf.distribute.Strategy.mirroredstrategy()
In the tensorflow docs tutorial I'm following, it is mentioned that you should call your train_step code from a function…
SumakuTension
- 542
- 1
- 9
- 26
0
votes
0 answers
RuntimeError: cublas runtime error : the GPU program failed to execute at /pytorch/aten/src/THC/THCBlas.cu:450
My program works properly with two GPUs but raise this error when reached to torch.bmm
energy = torch.bmm(proj_query,proj_key) # transpose check
RuntimeError: cublas runtime error : the GPU program failed to execute at…
Roqýah Ȝbđeen
- 1
- 2
0
votes
1 answer
How to use multiple GPUs on MATLAB - Out of memory on device
I have 4 GPU devices (11GB memory for each). Now, I want to add two big matrices A and B (6GB for each), so that it is not possible to have two matrices in the memory of one single device. I'd like to know how to perform the addition in this…
kaienfr
- 315
- 1
- 3
- 8
0
votes
0 answers
Model get stuck by using MirroredStrategy()
Im trying to use multiple GPU(2) in Keras, using MirroredStrategy() from Tensorflow. However it leads to the following error:
Epoch 1/5
WARNING:tensorflow:From…
Desperate Morty
- 66
- 2
- 11
0
votes
1 answer
TensorFlow on multiple GPU
Recently, I try to learn how to use Tensorflow on multiple GPU by reading the official tutorial. However, there is something that I am confused about. The following code is part of the official tutorial, which calculates the loss on single GPU.
def…
Sean
- 901
- 2
- 11
- 30
0
votes
1 answer
Using multiple gpus for regression
I am trying to write a tensorflow code for a regression problem. I have written the code and it works perfectly when run on a single gpu. However, I want to test how fast can it go a 2 gpus. I have modified the code for 2 gpus using tf.device. But I…
PalA
- 1
0
votes
1 answer
Tensorflow not allocate tensor/op to all available GPUs
I have two GPUs, a 6GB GTX Titan and a 11GB GTX 1080Ti.
Expectation: Tensorflow should automatically use all GPUs' memory.
Reality: Tensorflow map the two devices as gpu:0 and gpu:1, but it only use gpu:0 and never use gpu:1, when I increase the…
fuzhi
- 41
- 4
0
votes
1 answer
OpenCl wrong values when reading from multiple GPU
I have a kernel function that only writes number to a __global int* c
To be specific it looks like this:
__kernel void Add1(__global int* c)
{
*c = 3;
}
and in host code I have allocated memory for C value:
cl_mem bufferC[deviceNumber]; //…
Gzyniu
- 71
- 1
- 1
- 7
0
votes
1 answer
Multiple GPU code on Matlab runs for few seconds only
I am running the following MATLAB code on a system with one GTX 1080 and a K80 (with 2 GPUs)
delete(gcp('nocreate'));
parpool('local',2);
spmd
gpuDevice(labindex+1)
end
reset(gpuDevice(2))
reset(gpuDevice(3))
parfor i=1:100
…
user3275666
- 1
- 1
0
votes
1 answer
CUDA 5.0 context management with single application thread in multiple GPU environment
It seems that most tutorials, guides, books and Q&A from the web refers to CUDA 3 and 4.x, so that is why I'm asking it specifically about CUDA 5.0. To the question...
I would like to program for an environment with two CUDA devices, but use only…
Ricardo Inacio
- 171
- 10
0
votes
1 answer
How to dedicate DirectX to a GPU and dedicate CUDA to another GPU?
In my application, the GUI thread uses Direct2D to display images, allowing the user to pan and zoom, while another thread uses CUDA to do image processing. To prevent the image processing thread to be disturbed by the GUI thread and to speed things…
Papaya
- 332
- 3
- 15
0
votes
1 answer
OpenCL : Targeting Work-group to a specific device
Assuming that I have a multiprocessor machine. Can I bind my work-group to a specific device (processor) ?
Do we have any API to accomplish this task in openCL ?
Barun Parichha
- 166
- 4