Community
cancel
Showing results for 
Search instead for 
Did you mean: 
mvemp
Novice
1,242 Views

How is accelerator frequency decided in OpenCL designs?

If the opencl description contains multiple kernels connected with each other using channels, how is the frequency decided? Is it the kernel with least frequency? What about the frequency for isolated kernels (no communication to other kernels) in the same design? Does the frequncy depend upon the memory bandwidth of the FPGA board?

0 Kudos
2 Replies
KennyT_Intel
Moderator
73 Views

It would be whole system. you can refer to

 

https://www.intel.com/content/dam/www/programmable/us/en/pdfs/literature/hb/opencl-sdk/aocl_optimiza...

1-21 to check on your performance bottleneck.

 

Thanks

HRZ
Valued Contributor II
73 Views

A quick summary:

 

  • The same kernel clock is used for all kernels in the same CL file (which get placed and routed as one FPGA image). The operating frequency will be limited by the slowest kernel is the CL file. Whether the kernels are connected or not will not make a difference.
  • Kernel operating frequency is not affected the external memory bandwidth of the FPGA, but the external memory bandwidth can be affected by the kernel operating frequency.
Reply