Intel® DevCloud
Help for those needing help starting or connecting to the Intel® DevCloud
1638 Discussions

Iris Xe MAX on DevCloud

Cielo__Salvatore
Beginner
1,172 Views

Hi all, 

is it true that the new Xe GPUs are available on the DevCloud? I got the notification via email, and can't wait to test them for my computing kernels, but I couldn't find instructions on how to access them yet.

Thank you for any news!

Best, 

0 Kudos
1 Solution
JananiC_Intel
Moderator
1,148 Views

Hi,


Thanks for posting in Intel forums.


Intel Iris Xe MAX GPU's are available in devcloud now.You can connect the node with any of the below commands.

qsub -I -l nodes=1:dual_gpu:ppn=2 (or) qsub -I -l nodes=1:quad_gpu:ppn=2


As of now documentation is not updated.We have informed the concerned team about this.


Hope this helps!





View solution in original post

3 Replies
JananiC_Intel
Moderator
1,149 Views

Hi,


Thanks for posting in Intel forums.


Intel Iris Xe MAX GPU's are available in devcloud now.You can connect the node with any of the below commands.

qsub -I -l nodes=1:dual_gpu:ppn=2 (or) qsub -I -l nodes=1:quad_gpu:ppn=2


As of now documentation is not updated.We have informed the concerned team about this.


Hope this helps!





Cielo__Salvatore
Beginner
1,133 Views

Thanks a lot, I could access it.

oneMKL code gets easily a speedup of 1.5x over a gen9, on some simple gemm test.

I could not yet run general sycl code (some USM parallel_for), that works on gen9.
I got the following problem:

The program was built for 1 devices

Build program log for 'Intel(R) Gen12LP':

-11 (CL_BUILD_PROGRAM_FAILURE) 

terminate called after throwing an instance of 'cl::sycl::compile_program_error'

But no problem, I will wait for documentation.

Best, 

0 Kudos
JananiC_Intel
Moderator
1,125 Views

Hi,


Thanks for the update.


Since your initial query is resolved,we would discontinue monitoring this issue.Please raise a new thread if you have further issues.



0 Kudos
Reply