AN 831: Intel® FPGA SDK for OpenCL™: Host Pipelined Multithread

ID 683013
Date 11/20/2017
Public

1.4.4. Performance Evaluation

Intel has tested this framework on different benchmarks and by providing multiple input files. It was observed that the system throughput is same as the kernel throughput or even better.

The kernel throughput is greater than 3 GB/s for input files larger than 130 MB. This result determines that the design is very competitive for RTL design, and more powerful when compared to the CPU alone.