Intel GPU for 5G beamforming

[TestDR]

Hi,

We have a C++ beamforming application using Intel oneAPI Base tool kit and HPC tool kit. We want to offload matrix multiplication, EVD, matrix inversion to GPU. We use intel MKL gemm, LAPACK_cheevx,  getrf, getri at the moment and matrix sizes of 64 * 16 * 64.  Can you recommend a suitable GPU , first for Intel  Advisor offload modelling and later for buying ?

Thanks,

Rajitha

[Soumya_M_Intel]

Hi Rajitha,

Thanks for posting your concern on the forum. We are currently checking on the query with our resources.

Kindly wait for 2-3 business days. We will get back to you as soon as possible.

Hope you have a nice day ahead!

Regards,

Soumya

[Soumya_M_Intel]

Hi Rajitha,

We appreciate your patience. We are looking into your concern with our internal team. Please wait for 2-3 business days. We will get back to you as soon as we have an update.

Hope you have a nice day ahead!

Regards,

Soumya

[AlekhyaV_Intel]

Hi,

This forum is Analyzers forum which provides support to Intel Performance Profilers. We do not provide recommendations on the products.

For your reference, we have shared you all the Intel Advisor System requirements of the current versions (https://www.intel.com/content/dam/develop/external/us/en/documents/intel-advisor-2023-1-release-notes.pdf) You could find a list of supported GPUs.

If this resolves your issue, make sure to accept this as solution. This helps others with similar issues. Thank you!

Regards,

Alekhya

[AlekhyaV_Intel]

Hi,

We assume that the issue is resolved. If you need any further information, please post a new question as this thread will no longer be monitored by Intel.

Regards,

Alekhya