Abstract
Heterogeneous architectures featuring multiple hardware accelerators have been proposed as a promising solution for meeting the ever-increasing performance and power requirements of embedded systems. However, the existence of numerous design parameters may result in different architectural schemes and thus in extra design effort. To address this issue, OpenCL-based frameworks have been recently utilized for FPGA programming, to enable the portability of a source code to multiple architectures. However, such OpenCL frameworks focus on RTL design, thus not enabling rapid prototyping and abstracted modeling of complex systems. Virtual Prototyping aims to overcome this problem by enabling the system modeling in higher abstraction levels. This article combines the benefits of OpenCL and Virtual Prototyping, by proposing an OpenCL-based prototyping framework for data-parallel many-accelerator systems, which (a) creates a SystemC Virtual Platform from OpenCL, (b) provides a co-simulation environment for the host and the Virtual Platform, (c) offers memory and interconnection models for parallel data processing, and (d) enables the system evaluation with alternative real number representations (e.g., fixed-point or 16-bit floating-point).
- Altera OpenCL Guide. 2015. Altera SDK for OpenCL Optimization Guide. Retrieved from https://www.altera.com/content/dam/altera-www/global/en_US/pdfs/literature/hb/opencl-sdk/aocl_optimization_guide.pdf.Google Scholar
- AMD-OCL. 2017. AMD OpenCL. Retrieved from http://developer.amd.com/tools-and-sdks/opencl-zone/.Google Scholar
- Clang. 2017. Clang: A C language family frontend for LLVM. Retrieved from http://clang.llvm.org/.Google Scholar
- J. Cong, M. A. Ghodrat, M. Gill, B. Grigorian, Hui Huang, and G. Reinman. 2013. Composable accelerator-rich microprocessor enhanced for adaptivity and longevity. In Proceedings of the International Symposium on Low Power Electronics and Design (ISLPED’13). 305--310. Google ScholarDigital Library
- J. Cong, M. A. Ghodrat, M. Gill, B. Grigorian, and G. Reinman. 2012. Architecture support for accelerator-rich CMPs. In Proceedings of the 49th ACM/EDAC/IEEE Design Automation Conference (DAC’12). 843--849. Google ScholarDigital Library
- J. Cong, Mi. Gill, Y. Hao, G. Reinman, and B. Yuan. 2015. On-chip interconnection network for accelerator-rich architectures. In Proceedings of the 52nd Design Automation Conference (DAC’15). 8:1--8:6. Google ScholarDigital Library
- G. Deest, T. Yuki, O. Sentieys, and S. Derrien. 2014. Toward scalable source level accuracy analysis for floating-point to fixed-point conversion. In Proceedings of the 2014 IEEE/ACM International Conference on Computer-Aided Design (ICCAD’14). 726--733. Google ScholarDigital Library
- J. Diaz, C. Muñoz-Caro, and A. Niño. 2012. A survey of parallel programming models and tools in the multi and many-core era. IEEE Trans. Parallel Distrib. Syst. 23, 8 (Aug. 2012), 1369--1386. Google ScholarDigital Library
- HAPS. 2013. Synopsys High-performance ASIC Prototyping Systems. Retrieved from http://www.synopsys.com/Systems/FPGABasedPrototyping/Pages/HAPS.aspx.Google Scholar
- Intel-FPGA-OCL. 2017. Intel FPGA SDK for OpenCL. Retrieved from https://www.altera.com/products/design-software/embedded-software-developers/opencl/overview.html.Google Scholar
- ISSI DDR3. 2016. ISSI IS43/46TR16256A(L) DDR3 Specifications. Retrieved from http://www.issi.com/WW/pdf/43-46TR16256A-85120AL.pdf.Google Scholar
- K. Komatsu, K. Sato, Y. Arai, K. Koyama, H. Takizawa, and H. Kobayashi. 2010. Evaluating performance and portability of OpenCL programs. In Proceedings of the 5th International Workshop on Automatic Performance Tuning. 7--22.Google Scholar
- A. Mahzoon and B. Alizadeh. 2017. OptiFEX: A framework for exploring area-efficient floating point expressions on FPGAs with optimized exponent/mantissa widths. IEEE Trans. Very Large Scale Integr. Syst. 25, 1 (Jan. 2017), 198--209. Google ScholarDigital Library
- NVIDIA-OCL. 2017. NVIDIA OpenCL SDK. Retrieved from https://developer.nvidia.com/opencl.Google Scholar
- OpenCL. 2017a. OpenCL, by Khronos group. Retrieved from https://www.khronos.org/opencl/.Google Scholar
- OpenCL. 2017b. OpenCL Specifications 1.0. Retrieved from https://www.khronos.org/registry/cl/specs/opencl-1.0.pdf.Google Scholar
- OVP. 2017. Open Virtual Platforms website. Retrieved from http://www.ovpworld.org.Google Scholar
- PARADE. 2015. PARADE: Full-System Accelerator-Rich Architecture Simulator. Retrieved from http://vast.cs.ucla.edu/software/parade-ara-simulator.Google Scholar
- S. J. Parker and V. A. Chouliaras. 2016. An OpenCL software compilation framework targeting an SoC-FPGA VLIW chip multiprocessor. J. Syst. Architect. 68 (2016), 17--37. Google ScholarDigital Library
- Rodinia. 2016. Rodinia: A Benchmark Suite for Heterogeneous Computing. Retrieved from http://lava.cs.virginia.edu/Rodinia/.Google Scholar
- SDAccel. 2017. Xilinx SDAccel. Retrieved from www.xilinx.com/products/design-tools/software-zone/sdaccel.html.Google Scholar
- SDAccel Guide. 2016. Xilinx SDAccel Development Environment (UG1023). Retrieved from http://www.xilinx.com/support/documentation/sw_man uals/xilinx2015_1/ug1023-sdaccel-user-guide.pdf.Google Scholar
- E. Sotiriou-Xanthopoulos, L. Masing, K. Siozios, G. Economakos, D. Soudris, and J. Becker. 2016b. An OpenCL-based framework for rapid virtual prototyping of heterogeneous architectures. In Proceedings of the Conference on Video and Image Processing Emulation System (ViPES’16), Co-located with the International Conference on Embedded Computer Systems (IC-SAMOS’16). 372--377.Google Scholar
- E. Sotiriou-Xanthopoulos, S. Xydis, K. Siozios, and G. Economakos. 2015. A virtual platform for exploring hierarchical interconnection for many-accelerator systems. In Proceedings of the International Conference on Embedded Computer Systems (IC-SAMOS’15). 384--389.Google Scholar
- E. Sotiriou-Xanthopoulos, S. Xydis, K. Siozios, G. Economakos, and D. Soudris. 2014. Effective platform-level exploration for heterogeneous multicores exploiting simulation-induced slacks. In Proceedings of the Workshop on Parallel Programming and Run-Time Management Techniques for Many-core Architectures (PARMA-DITAM’14). ACM, New York, NY, Article 13, 4 pages. Google ScholarDigital Library
- E. Sotiriou-Xanthopoulos, S. Xydis, K. Siozios, G. Economakos, and D. Soudris. 2016a. An integrated exploration and virtual platform framework for many-accelerator heterogeneous systems. ACM Trans. Embed. Comput. Syst. 15, 3, Article 43 (2016a), 26 pages. Google ScholarDigital Library
- SystemC. 2017. SystemC official website. Retrieved from http://www.accellera.org/downloads/standards/systemc.Google Scholar
- Terasic. 2017. Terasic website. Retrieved from http://www.terasic.com.tw.Google Scholar
- Vista Virtual Prototyping. 2017. Vista Virtual Prototyping, by Mentor Graphics. Retrieved from https://www.mentor.com/esl/vista/virtual-prototyping/.Google Scholar
- Vivado-UserGuide. 2015. Vivado User Guide, UG902 (v2015.4). Retrieved from http://www.xilinx.com/support/documentation/sw_man uals/xilinx2015_4/ug902-vivado-high-level-synthesis.pdf.Google Scholar
- J. van der Zijp. 2012. Fast half float conversions. Retrieved from ftp://www.fox-toolkit.org/pub/fasthalffloatconversion.pdf.Google Scholar
- Zynq. 2017. Xilinx Zynq-7000 SoCs. Retrieved from http://www.xilinx.com/products/boards-and-kits/device-family/nav-zynq-7000.html.Google Scholar
Index Terms
- OpenCL-based Virtual Prototyping and Simulation of Many-Accelerator Architectures
Recommendations
Co-simulation framework of SystemC SoC virtual prototype and custom logic (abstract only)
FPGA '13: Proceedings of the ACM/SIGDA international symposium on Field programmable gate arraysTo address the increasing demand of System-on-Chip (SoC) for high performance applications and IP programmability, specialized SoC with custom logic is developed in a single chip or multi-chip system. Like any other SoC platforms, early software ...
Nuclear Reactor Simulations on OpenCL FPGA Platform
FPGA '19: Proceedings of the 2019 ACM/SIGDA International Symposium on Field-Programmable Gate ArraysField-programmable gate arrays (FPGAs) are becoming a promising choice as a heterogeneous computing component for scientific computing when floating-point optimized architectures are added to the current FPGAs. The maturing high-level synthesis (HLS) ...
Nuclear Reactor Simulation on OpenCL FPGA: a Case Study of RSBench
IWOCL '18: Proceedings of the International Workshop on OpenCLField-programmable gate arrays (FPGAs) are becoming a promising choice as a heterogeneous computing component for scientific computing when floating-point optimized architectures are added to the current FPGAs. The emerging high-level synthesis tools ...
Comments