How to confirm that my GPU is running at full capacity? The GPU load is so low
My environment: windows10, tensorflow-gpu, single Nvidia RTX2070 (8GB memory), using tflearn to run AlexNet with the OxFlowers17 data set, batch_size=128.
I got this information:enter image description here
This means the GPU is running, right?
However, when I check the GPU usage with GPU-Z, the GPU load is low (The code is doing training, epoch information is being printed in the output window). Does this mean that my GPU is not running at full capacity?enter image description here
After I closed PyCharm and with no big program running on my computer, the GPU usage is this:enter image description here
Then I have three questions:
The GPU load is low. Does this mean the GPU is not running at full capacity?
I forgot to set the gpu_memory_fraction. If I want to allocate a suitable gpu memory, how much should I set? (e.g. AlexNet with OxFlowers17 data set, the size of the data set is about 60MB.)
Why was the System Memory Used also so high? It has already used all the free gpu memory.
PS: This is the CPU (i5-8500) information when training:enter image description here
python tensorflow gpu
asked Nov 24 '18 at 2:54
feiyuhuahuofeiyuhuahuo
165
add a comment |
My environment: windows10, tensorflow-gpu, single Nvidia RTX2070 (8GB memory), using tflearn to run AlexNet with the OxFlowers17 data set, batch_size=128.
I got this information:enter image description here
This means the GPU is running, right?
However, when I check the GPU usage with GPU-Z, the GPU load is low (The code is doing training, epoch information is being printed in the output window). Does this mean that my GPU is not running at full capacity?enter image description here
After I closed PyCharm and with no big program running on my computer, the GPU usage is this:enter image description here
Then I have three questions:
The GPU load is low. Does this mean the GPU is not running at full capacity?
I forgot to set the gpu_memory_fraction. If I want to allocate a suitable gpu memory, how much should I set? (e.g. AlexNet with OxFlowers17 data set, the size of the data set is about 60MB.)
Why was the System Memory Used also so high? It has already used all the free gpu memory.
PS: This is the CPU (i5-8500) information when training:enter image description here
python tensorflow gpu
asked Nov 24 '18 at 2:54
feiyuhuahuofeiyuhuahuo
165
1
Based on the stats, your GPU seems to be running, but there might be another bottleneck. What does CPU load look like?
– user8408080
Nov 24 '18 at 3:00
is the code doing inference? if so then there is nothing to worry about, if you are training, gpu load should increase. Tensorflow by default allocate all GPU memory, so once you set the gpu_memory_fraction, it will reduce per your spec.
– teng
Nov 24 '18 at 3:04
@user8408080: I have added the CPU information. Is there any problem then?
– feiyuhuahuo
Nov 24 '18 at 3:30
@teng: My code is training, but the GPU load is still so low. I edited my question again.
– feiyuhuahuo
Nov 24 '18 at 3:32
training data size is just 60MB... i thought this was a typo, but looks like it is just a small dataset, robots.ox.ac.uk/~vgg/data/flowers/17 I don't think you need to be worried, you are just using a relatively good GPU that's all. If you still are not convinced, try varying the batch size to see if the gpu load changes.
– teng
Nov 24 '18 at 3:37
add a comment |
My environment: windows10, tensorflow-gpu, single Nvidia RTX2070 (8GB memory), using tflearn to run AlexNet with the OxFlowers17 data set, batch_size=128.
I got this information:enter image description here
This means the GPU is running, right?
However, when I check the GPU usage with GPU-Z, the GPU load is low (The code is doing training, epoch information is being printed in the output window). Does this mean that my GPU is not running at full capacity?enter image description here
After I closed PyCharm and with no big program running on my computer, the GPU usage is this:enter image description here
Then I have three questions:
The GPU load is low. Does this mean the GPU is not running at full capacity?
I forgot to set the gpu_memory_fraction. If I want to allocate a suitable gpu memory, how much should I set? (e.g. AlexNet with OxFlowers17 data set, the size of the data set is about 60MB.)
Why was the System Memory Used also so high? It has already used all the free gpu memory.
PS: This is the CPU (i5-8500) information when training:enter image description here
python tensorflow gpu
asked Nov 24 '18 at 2:54
feiyuhuahuofeiyuhuahuo
165
My environment: windows10, tensorflow-gpu, single Nvidia RTX2070 (8GB memory), using tflearn to run AlexNet with the OxFlowers17 data set, batch_size=128.
I got this information:enter image description here
This means the GPU is running, right?
However, when I check the GPU usage with GPU-Z, the GPU load is low (The code is doing training, epoch information is being printed in the output window). Does this mean that my GPU is not running at full capacity?enter image description here
After I closed PyCharm and with no big program running on my computer, the GPU usage is this:enter image description here
Then I have three questions:
The GPU load is low. Does this mean the GPU is not running at full capacity?
I forgot to set the gpu_memory_fraction. If I want to allocate a suitable gpu memory, how much should I set? (e.g. AlexNet with OxFlowers17 data set, the size of the data set is about 60MB.)
Why was the System Memory Used also so high? It has already used all the free gpu memory.
PS: This is the CPU (i5-8500) information when training:enter image description here
python tensorflow gpu
python tensorflow gpu
asked Nov 24 '18 at 2:54
feiyuhuahuofeiyuhuahuo
165
asked Nov 24 '18 at 2:54
feiyuhuahuofeiyuhuahuo
165
edited Nov 24 '18 at 3:36
feiyuhuahuo
asked Nov 24 '18 at 2:54
feiyuhuahuofeiyuhuahuo
165
asked Nov 24 '18 at 2:54
feiyuhuahuofeiyuhuahuo
165
asked Nov 24 '18 at 2:54
feiyuhuahuofeiyuhuahuo
165
165
1
Based on the stats, your GPU seems to be running, but there might be another bottleneck. What does CPU load look like?
– user8408080
Nov 24 '18 at 3:00
is the code doing inference? if so then there is nothing to worry about, if you are training, gpu load should increase. Tensorflow by default allocate all GPU memory, so once you set the gpu_memory_fraction, it will reduce per your spec.
– teng
Nov 24 '18 at 3:04
@user8408080: I have added the CPU information. Is there any problem then?
– feiyuhuahuo
Nov 24 '18 at 3:30
@teng: My code is training, but the GPU load is still so low. I edited my question again.
– feiyuhuahuo
Nov 24 '18 at 3:32
training data size is just 60MB... i thought this was a typo, but looks like it is just a small dataset, robots.ox.ac.uk/~vgg/data/flowers/17 I don't think you need to be worried, you are just using a relatively good GPU that's all. If you still are not convinced, try varying the batch size to see if the gpu load changes.
– teng
Nov 24 '18 at 3:37
add a comment |
1
Based on the stats, your GPU seems to be running, but there might be another bottleneck. What does CPU load look like?
– user8408080
Nov 24 '18 at 3:00
is the code doing inference? if so then there is nothing to worry about, if you are training, gpu load should increase. Tensorflow by default allocate all GPU memory, so once you set the gpu_memory_fraction, it will reduce per your spec.
– teng
Nov 24 '18 at 3:04
@user8408080: I have added the CPU information. Is there any problem then?
– feiyuhuahuo
Nov 24 '18 at 3:30
@teng: My code is training, but the GPU load is still so low. I edited my question again.
– feiyuhuahuo
Nov 24 '18 at 3:32
training data size is just 60MB... i thought this was a typo, but looks like it is just a small dataset, robots.ox.ac.uk/~vgg/data/flowers/17 I don't think you need to be worried, you are just using a relatively good GPU that's all. If you still are not convinced, try varying the batch size to see if the gpu load changes.
– teng
Nov 24 '18 at 3:37
1
1
Based on the stats, your GPU seems to be running, but there might be another bottleneck. What does CPU load look like?
– user8408080
Nov 24 '18 at 3:00
Based on the stats, your GPU seems to be running, but there might be another bottleneck. What does CPU load look like?
– user8408080
Nov 24 '18 at 3:00
is the code doing inference? if so then there is nothing to worry about, if you are training, gpu load should increase. Tensorflow by default allocate all GPU memory, so once you set the gpu_memory_fraction, it will reduce per your spec.
– teng
Nov 24 '18 at 3:04
is the code doing inference? if so then there is nothing to worry about, if you are training, gpu load should increase. Tensorflow by default allocate all GPU memory, so once you set the gpu_memory_fraction, it will reduce per your spec.
– teng
Nov 24 '18 at 3:04
@user8408080: I have added the CPU information. Is there any problem then?
– feiyuhuahuo
Nov 24 '18 at 3:30
@user8408080: I have added the CPU information. Is there any problem then?
– feiyuhuahuo
Nov 24 '18 at 3:30
@teng: My code is training, but the GPU load is still so low. I edited my question again.
– feiyuhuahuo
Nov 24 '18 at 3:32
@teng: My code is training, but the GPU load is still so low. I edited my question again.
– feiyuhuahuo
Nov 24 '18 at 3:32
training data size is just 60MB... i thought this was a typo, but looks like it is just a small dataset, robots.ox.ac.uk/~vgg/data/flowers/17 I don't think you need to be worried, you are just using a relatively good GPU that's all. If you still are not convinced, try varying the batch size to see if the gpu load changes.
– teng
Nov 24 '18 at 3:37
training data size is just 60MB... i thought this was a typo, but looks like it is just a small dataset, robots.ox.ac.uk/~vgg/data/flowers/17 I don't think you need to be worried, you are just using a relatively good GPU that's all. If you still are not convinced, try varying the batch size to see if the gpu load changes.
– teng
Nov 24 '18 at 3:37
add a comment |
0
active
oldest
votes
Your Answer
StackExchange.ifUsing("editor", function () {
StackExchange.using("externalEditor", function () {
StackExchange.using("snippets", function () {
StackExchange.snippets.init();
});
});
}, "code-snippets");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "1"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53454781%2fhow-to-confirm-that-my-gpu-is-running-at-full-capacity-the-gpu-load-is-so-low%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
0
active
oldest
votes
0
active
oldest
votes
active
oldest
votes
active
oldest
votes
Thanks for contributing an answer to Stack Overflow!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53454781%2fhow-to-confirm-that-my-gpu-is-running-at-full-capacity-the-gpu-load-is-so-low%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
1
Based on the stats, your GPU seems to be running, but there might be another bottleneck. What does CPU load look like?
– user8408080
Nov 24 '18 at 3:00
is the code doing inference? if so then there is nothing to worry about, if you are training, gpu load should increase. Tensorflow by default allocate all GPU memory, so once you set the gpu_memory_fraction, it will reduce per your spec.
– teng
Nov 24 '18 at 3:04
@user8408080: I have added the CPU information. Is there any problem then?
– feiyuhuahuo
Nov 24 '18 at 3:30
@teng: My code is training, but the GPU load is still so low. I edited my question again.
– feiyuhuahuo
Nov 24 '18 at 3:32
training data size is just 60MB... i thought this was a typo, but looks like it is just a small dataset, robots.ox.ac.uk/~vgg/data/flowers/17 I don't think you need to be worried, you are just using a relatively good GPU that's all. If you still are not convinced, try varying the batch size to see if the gpu load changes.
– teng
Nov 24 '18 at 3:37