Abstract
The rapid increase in FPGA logic capacity has enabled the prototyping of multiprocessor Network-on-Chip (NoC) architectures. However, the design space exploration of these complex architectures is highly time consuming with traditional methodologies for FPGA design.
Our paper addresses the challenges of multiprocessor network design with the Blueshell framework for generating multiprocessor networks on chip (NoC) and a coupled Java software stack, Network-Chi. With Blueshell hardware is mconstructed from high-level components including processors and routers using concise Bluespec System Verilog. The Network-Chi software framework is also presented to enable programming the on-chip processors in a familiar Java style and without exposing the low-level systems programming to the application designer.
We demonstrate that Blueshell systems with as many as 20 processors can be implemented on a modestly sized FPGA. Performance figures for a selection of distributed applications are also provided.
- P. Abad, P. Prieto, L. G. Menezo, A. Colaso, V. Puente, and J.-A. Gregorio. TOPAZ: An Open-Source Interconnection Network Simulator for Chip Multiprocessors and Supercomputers. 2012 IEEE/ACM Sixth International Symposium on Networks-on-Chip, pages 99--106, May 2012. Google ScholarDigital Library
- J. Auerbach, D. Bacon, and I. Burcea. A compiler and runtime for heterogeneous computing. DAC '12 Proceedings of the 49th Annual Design Automation Conference, (1):271, 2012. Application 2x1 3x3 5x4 JVM 3x3 MD5 3,330 12,300 11,200 7,630,000 MD5 (-unsafe) 8,680 42,900 50,700 - Table 6: MD5 Benchmark performance across various network sizes. The values are the mean count of MD5 hashes calculated per second. Higher is better. Google ScholarDigital Library
- Bluespec Inc. Bluespec System Verilog (BSV) URL:http://www.bluespec.com/products, 2013.Google Scholar
- E. A. Carara, R. P. de Oliveira, N. L. V. Calazans, and F. G. Moraes. HeMPS - a framework for NoC-based MPSoC generation. In 2009 IEEE International Symposium on Circuits and Systems, pages 1345--1348. IEEE, May 2009.Google ScholarCross Ref
- M. Dev Gomony, B. Akesson, and K. Goossens. Architecture and Optimal Configuration of a Real-Time Multi-Channel Memory Controller. In Proc. Design, Automation and Test in Europe Conference and Exhibition (DATE), 2013. Google ScholarDigital Library
- E. Fernandez-Alonso and D. Castells-Rufas. Survey of NoC and Programming Models Proposals for MPSoC. IJCSI International Journal of Computer Science Issues,, 9(2):22--32, 2012.Google Scholar
- GAPH - PUCRS. Atlas NoC Framework URL:https://corfu.pucrs.br/redmine/projects/atlas, 2011.Google Scholar
- S. Huang, A. Hormati, D. Bacon, and R. Rabbah. Liquid metal: Object-oriented programming across the hardware/software boundary. ECOOP '08 - Proceedings of the 22nd European conference on Object-Oriented Programming, pages 76--103, 2008. Google ScholarDigital Library
- G. Kahn. The semantics of a simple language for parallel programming. In J. L. Rosenfeld, editor, Information processing, pages 471--475, Stockholm, Sweden, 1974. North Holland, Amsterdam.Google Scholar
- F. Moraes, N. Calazans, A. Mello, L. Möller, and L. Ost. HERMES: an infrastructure for low area overhead packet-switching networks on chip. Integration, the VLSI Journal, 38(1):69--93, Oct. 2004. Google ScholarDigital Library
- M. K. Papamichael and J. C. Hoe. CONNECT: re-examining conventional wisdom for designing nocs in the context of FPGAs. In Proceedings of the ACM/SIGDA international symposium on Field Programmable Gate Arrays, FPGA '12, pages 37--46, New York, NY, USA, 2012. ACM. Google ScholarDigital Library
- G. Plumbridge and N. Audsley. Translating Java for Resource Constrained Embedded Systems. 7th International Workshop on Reconfigurable and Communication-Centric Systems-on-Chip (ReCoSoC), pages 1--8, July 2012.Google Scholar
- S. Rhoads. Plasma :: most MIPS I(TM) opcodes: OpenCores URL:http://opencores.org/project,plasma, 2013.Google Scholar
- U. P. Schultz, K. Burgaard, F. G. Christensen, and J. r. L. Knudsen. Compiling java for low-end embedded systems. LCTES '03 Proceedings of the 2003 ACM SIGPLAN conference on Language, compiler, and tool for embedded systems, 38(7):42, July 2003. Google ScholarDigital Library
- A. Varma and S. S. Bhattacharyya. Java-through-C Compilation: An Enabling Technology for Java in Embedded Systems. DATE '04 Proceedings of the conference on Design, automation and test in Europe - Volume 3, page 30161, Feb. 2004. Google ScholarDigital Library
- D. Wang, N. Jerger, and J. Steffan. DART: A programmable architecture for NoC simulation on FPGAs. Networks on Chip (NoCS), 2011 Fifth IEEE/ACM International Symposium on, 2011. Google ScholarDigital Library
- Xilinx. AXI Reference Guide UG761 (v14.3). 2012.Google Scholar
- Xilinx. MicroBlaze Processor Reference Guide UG081 (v14.2). Technical report, Xilinx Inc., 2012Google Scholar
Index Terms
- Blueshell: a platform for rapid prototyping of multiprocessor NoCs and accelerators
Recommendations
Synthesizable Standard Cell FPGA Fabrics Targetable by the Verilog-to-Routing CAD Flow
Special Section on Field Programmable Logic and Applications 2015 and Regular PapersIn this article, we consider implementing field-programmable gate arrays (FPGAs) using a standard cell design methodology and present a framework for the automated generation of synthesizable FPGA fabrics. The open-source Verilog-to-Routing (VTR) FPGA ...
Comments