What is the location of output of thrust::reduce operation(GPU RAM or CPU RAM)? [duplicate]
This question already has an answer here:
thrust reduction result on device memory
2 answers
When we use thrust::reduce in the following example, the output is an int. Is this output (sum variable in code) located on the GPU or CPU RAM?
If it is in CPU RAM, how to access/retain the variable in the gpu?
The reduce operation happens on the device(GPU) so at some point the output should be in the GPU.
#include <thrust/reduce.h>
#include <thrust/execution_policy.h>
#include <thrust/device_vector.h>
int main()
{
thrust::device_vector<int> D(6);
D[0]=0;
D[1]=1;
D[2]=2;
D[3]=3;
D[4]=4;
D[5]=5;
int sum = thrust::reduce(thrust::device,D.begin(), D.end(), (int) 0, thrust::plus<int>());
}
cuda reduce ram thrust
asked Nov 21 '18 at 10:10
user27665user27665
357415
marked as duplicate by talonmies cuda
Users with the cuda badge can single-handedly close cuda questions as duplicates and reopen them as needed.
StackExchange.ready(function() {
if (StackExchange.options.isMobile) return;
$('.dupe-hammer-message-hover:not(.hover-bound)').each(function() {
var $hover = $(this).addClass('hover-bound'),
$msg = $hover.siblings('.dupe-hammer-message');
$hover.hover(
function() {
$hover.showInfoMessage('', {
messageElement: $msg.clone().show(),
transient: false,
position: { my: 'bottom left', at: 'top center', offsetTop: -7 },
dismissable: false,
relativeToBody: true
});
},
function() {
StackExchange.helpers.removeMessages();
}
);
});
});
Nov 21 '18 at 16:47
This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.
add a comment |
This question already has an answer here:
thrust reduction result on device memory
2 answers
When we use thrust::reduce in the following example, the output is an int. Is this output (sum variable in code) located on the GPU or CPU RAM?
If it is in CPU RAM, how to access/retain the variable in the gpu?
The reduce operation happens on the device(GPU) so at some point the output should be in the GPU.
#include <thrust/reduce.h>
#include <thrust/execution_policy.h>
#include <thrust/device_vector.h>
int main()
{
thrust::device_vector<int> D(6);
D[0]=0;
D[1]=1;
D[2]=2;
D[3]=3;
D[4]=4;
D[5]=5;
int sum = thrust::reduce(thrust::device,D.begin(), D.end(), (int) 0, thrust::plus<int>());
}
cuda reduce ram thrust
asked Nov 21 '18 at 10:10
user27665user27665
357415
marked as duplicate by talonmies cuda
Users with the cuda badge can single-handedly close cuda questions as duplicates and reopen them as needed.
StackExchange.ready(function() {
if (StackExchange.options.isMobile) return;
$('.dupe-hammer-message-hover:not(.hover-bound)').each(function() {
var $hover = $(this).addClass('hover-bound'),
$msg = $hover.siblings('.dupe-hammer-message');
$hover.hover(
function() {
$hover.showInfoMessage('', {
messageElement: $msg.clone().show(),
transient: false,
position: { my: 'bottom left', at: 'top center', offsetTop: -7 },
dismissable: false,
relativeToBody: true
});
},
function() {
StackExchange.helpers.removeMessages();
}
);
});
});
Nov 21 '18 at 16:47
This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.
1
Please check this answer by Robert Crovella
– sgarizvi
Nov 21 '18 at 10:44
add a comment |
This question already has an answer here:
thrust reduction result on device memory
2 answers
When we use thrust::reduce in the following example, the output is an int. Is this output (sum variable in code) located on the GPU or CPU RAM?
If it is in CPU RAM, how to access/retain the variable in the gpu?
The reduce operation happens on the device(GPU) so at some point the output should be in the GPU.
#include <thrust/reduce.h>
#include <thrust/execution_policy.h>
#include <thrust/device_vector.h>
int main()
{
thrust::device_vector<int> D(6);
D[0]=0;
D[1]=1;
D[2]=2;
D[3]=3;
D[4]=4;
D[5]=5;
int sum = thrust::reduce(thrust::device,D.begin(), D.end(), (int) 0, thrust::plus<int>());
}
cuda reduce ram thrust
asked Nov 21 '18 at 10:10
user27665user27665
357415
This question already has an answer here:
thrust reduction result on device memory
2 answers
When we use thrust::reduce in the following example, the output is an int. Is this output (sum variable in code) located on the GPU or CPU RAM?
If it is in CPU RAM, how to access/retain the variable in the gpu?
The reduce operation happens on the device(GPU) so at some point the output should be in the GPU.
#include <thrust/reduce.h>
#include <thrust/execution_policy.h>
#include <thrust/device_vector.h>
int main()
{
thrust::device_vector<int> D(6);
D[0]=0;
D[1]=1;
D[2]=2;
D[3]=3;
D[4]=4;
D[5]=5;
int sum = thrust::reduce(thrust::device,D.begin(), D.end(), (int) 0, thrust::plus<int>());
}
This question already has an answer here:
thrust reduction result on device memory
2 answers
cuda reduce ram thrust
cuda reduce ram thrust
asked Nov 21 '18 at 10:10
user27665user27665
357415
asked Nov 21 '18 at 10:10
user27665user27665
357415
asked Nov 21 '18 at 10:10
user27665user27665
357415
asked Nov 21 '18 at 10:10
user27665user27665
357415
asked Nov 21 '18 at 10:10
user27665user27665
357415
357415
marked as duplicate by talonmies cuda
Users with the cuda badge can single-handedly close cuda questions as duplicates and reopen them as needed.
StackExchange.ready(function() {
if (StackExchange.options.isMobile) return;
$('.dupe-hammer-message-hover:not(.hover-bound)').each(function() {
var $hover = $(this).addClass('hover-bound'),
$msg = $hover.siblings('.dupe-hammer-message');
$hover.hover(
function() {
$hover.showInfoMessage('', {
messageElement: $msg.clone().show(),
transient: false,
position: { my: 'bottom left', at: 'top center', offsetTop: -7 },
dismissable: false,
relativeToBody: true
});
},
function() {
StackExchange.helpers.removeMessages();
}
);
});
});
Nov 21 '18 at 16:47
This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.
marked as duplicate by talonmies cuda
Users with the cuda badge can single-handedly close cuda questions as duplicates and reopen them as needed.
StackExchange.ready(function() {
if (StackExchange.options.isMobile) return;
$('.dupe-hammer-message-hover:not(.hover-bound)').each(function() {
var $hover = $(this).addClass('hover-bound'),
$msg = $hover.siblings('.dupe-hammer-message');
$hover.hover(
function() {
$hover.showInfoMessage('', {
messageElement: $msg.clone().show(),
transient: false,
position: { my: 'bottom left', at: 'top center', offsetTop: -7 },
dismissable: false,
relativeToBody: true
});
},
function() {
StackExchange.helpers.removeMessages();
}
);
});
});
Nov 21 '18 at 16:47
This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.
1
Please check this answer by Robert Crovella
– sgarizvi
Nov 21 '18 at 10:44
add a comment |
1
Please check this answer by Robert Crovella
– sgarizvi
Nov 21 '18 at 10:44
1
1
Please check this answer by Robert Crovella
– sgarizvi
Nov 21 '18 at 10:44
Please check this answer by Robert Crovella
– sgarizvi
Nov 21 '18 at 10:44
add a comment |
1 Answer
1
active
oldest
votes
The result sum is on the CPU, you can refer to this documentation.
The result was probably at some point on the GPU, if the final result was computed on it. I didn't checked the implementation, but it could be possible that the final result is computed on the CPU. If you need to access it on the GPU, just pass it as a kernel argument, or copy it into global memory.
answered Nov 21 '18 at 10:43
Robin ThoniRobin Thoni
949720
add a comment |
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
The result sum is on the CPU, you can refer to this documentation.
The result was probably at some point on the GPU, if the final result was computed on it. I didn't checked the implementation, but it could be possible that the final result is computed on the CPU. If you need to access it on the GPU, just pass it as a kernel argument, or copy it into global memory.
answered Nov 21 '18 at 10:43
Robin ThoniRobin Thoni
949720
add a comment |
The result sum is on the CPU, you can refer to this documentation.
The result was probably at some point on the GPU, if the final result was computed on it. I didn't checked the implementation, but it could be possible that the final result is computed on the CPU. If you need to access it on the GPU, just pass it as a kernel argument, or copy it into global memory.
answered Nov 21 '18 at 10:43
Robin ThoniRobin Thoni
949720
add a comment |
The result sum is on the CPU, you can refer to this documentation.
The result was probably at some point on the GPU, if the final result was computed on it. I didn't checked the implementation, but it could be possible that the final result is computed on the CPU. If you need to access it on the GPU, just pass it as a kernel argument, or copy it into global memory.
answered Nov 21 '18 at 10:43
Robin ThoniRobin Thoni
949720
The result sum is on the CPU, you can refer to this documentation.
The result was probably at some point on the GPU, if the final result was computed on it. I didn't checked the implementation, but it could be possible that the final result is computed on the CPU. If you need to access it on the GPU, just pass it as a kernel argument, or copy it into global memory.
answered Nov 21 '18 at 10:43
Robin ThoniRobin Thoni
949720
answered Nov 21 '18 at 10:43
Robin ThoniRobin Thoni
949720
answered Nov 21 '18 at 10:43
Robin ThoniRobin Thoni
949720
answered Nov 21 '18 at 10:43
Robin ThoniRobin Thoni
949720
949720
add a comment |
add a comment |
1
Please check this answer by Robert Crovella
– sgarizvi
Nov 21 '18 at 10:44