FPGA-based Implementation of Memory-Intensive Application using OpenCL

ABSTRACT

Systems with heterogeneous architectures, such as field programmable gate arrays (FPGAs) and graphics processing units (GPUs), are expected to boost the throughput in high-performance computing applications. However, the user may struggle to write sophisticated programs that are performed effectively on complicated heterogeneous systems. High-level synthesis (HLS) is a solution that can reduce the development time. This study focuses on the implementation of OpenCL programming as a type of HLS design on FPGA boards. The Himeno benchmark, which is a suitable benchmark for the measurement of memory-intensive applications, is chosen as a verification program. We found that the OpenCL-based implementation achieves reasonable performance on FPGAs by demonstrating the implementation of temporal blocking combined with shift register implementation simultaneously. For Stratix V DE5-Net FPGA, our current implementation achieves 10.62 GFLOPS, or 75% of the theoretical performance. Meanwhile, for Arria 10 A10PL4 FPGA, the peak performance reaches 13.95 GFLOPS, or 76% of the theoretical performance.

References

1. Naruse, A., Sumimoto, S., Kumon, K. Acceleration Technique of Computational Fluid Dynanics on GPGPU - Over 60 GFLOPS Himeno Benchmark Performance on 1 GPU. IPSJ SIG Technical Reports, no. 99, pp. 49--54, 2008.
2. Matsuoka et al, A. GPU accelerated computingâĂŞfrom hype to mainstream, the rebirth of vector computing. Journal of Physics, vol. 180, no. 1, 2009.
3. Phillips, E.H., Fatica, M. Implementing the Himeno benchmark with CUDA on GPU clusters. IEEE Intl. Symposium on Parallel Distributed Processing, 2010.
4. Sato, Y., Inoguchi, Y., Luk, W., Nakamura, T. Evaluating reconfigurable dataflow computing using the Himeno benchmark. International Conference on Reconfigurable Computing and FPGAs, 2012.
5. Korch, M., Rauber, T., Scholtes, C. Memory-Intensive Applications on a Many-Core Processor. IEEE International Conference on High Performance Computing and Communications, 2011.
6. Kobayashi, H. Feasibility Study of a Future HPC System for Memory-Intensive Applications. Sustained Simulation Performance. Springer, Cham, 2014.
7. Himeno Benchmark, http://accc.riken.jp/.
8. Waidyasooriya, H.M., Hariyama, M., Uchiyama, K.. Design of FPGA-Based Computing Systems with OpenCL. Springer International Publishing AG 2018.
9. Jia, Q., Zhou, H. Tuning Stencil codes in OpenCL for FPGAs. IEEE Intl. Conference on Computer Design, 2016.
10. Waidyasooriya, H.M et al, OpenCL-Based FPGA-Platform for Stencil Computation and Its Optimization Methodology. IEEE Trans. on Parallel and Distributed Systems, 2016.
11. Zohouri, H.R., Podobas, A., Matsuoka, S., Combined Spatial and Temporal Blocking for High-Performance Stencil Computation on FPGAs Using OpenCL. International Symposium on Field-Programmable Gate Arrays, 2018.