Is it possible to launch multiple kernels on multiple GPUs concurrently from a single thread in cuda 4.0?
Is it possible to launch multiple kernels on multiple GPUs concurrently from a single
Share
Sign Up to our social questions and Answers Engine to ask questions, answer people’s questions, and connect with other people.
Login to our social questions & Answers Engine to ask questions answer people’s questions & connect with other people.
Lost your password? Please enter your email address. You will receive a link and will create a new password via email.
Please briefly explain why you feel this question should be reported.
Please briefly explain why you feel this answer should be reported.
Please briefly explain why you feel this user should be reported.
Is it possible to launch multiple kernels on multiple GPUs concurrently from a single thread in cuda 4.0?
To use multiple GPUs from a single thread, you can switch between cuda contexts (each of which is bound is bound to a GPU) and launch kernels asynchronously. In effect you will be running multiple kernels across multiple GPUs this way.
However if you have cards with compute capability > 2.0, you can also run kernels concurrently as shown in the comments above. You can find the post about concurrent kernel execution over here.
Ofcourse you can use both if you have multiple cards with compute capability >= 2.0.