Abstract
High-Performance Reconfigurable Computers (HPRCs) consist of one or more standard microprocessors tightly-coupled with one or more reconfigurable FPGAs. HPRCs have been shown to provide good speedups and good cost/performance ratios, but not necessarily ease of use, leading to a slow acceptance of this technology. HPRCs introduce new design challenges, such as the lack of portability across platforms, incompatibilities with legacy code, users reluctant to change their code base, a prolonged learning curve, and the need for a system-level Hardware/Software co-design development flow. This article presents the evolution and current work on TMD-MPI, which started as an MPI-based programming model for Multiprocessor Systems-on-Chip implemented in FPGAs, and has now evolved to include multiple X86 processors. TMD-MPI is shown to address current design challenges in HPRC usage, suggesting that the MPI standard has enough syntax and semantics to program these new types of parallel architectures. Also presented is the TMD-MPI Ecosystem, which consists of research projects and tools that are developed around TMD-MPI to further improve HPRC usability. Finally, we present preliminary communication performance measurements.
- Aggarwal, V., Troxel, I. A., and George, A. D. 2004. Design and analysis of parallel N-queens on reconfigurable hardware with Handel-C and MPI. In Proceedings of the MAPLD International Conference.Google Scholar
- Amirix Systems, Inc. 2009. http://www.amirix.com/ {Accessed 3/09}.Google Scholar
- Callahan, D., Chamberlain, B., and Zima, H. 2004. The cascade high productivity language. In Proceedings of the 9th International Workshop on High-Level Parallel Programming Models and Supportive Environments. 52--60.Google Scholar
- Chang, C., Wawrzynek, J., and Brodersen, R. W. 2005. BEE2: A high-end reconfigurable computing system. IEEE Des. Test 22, 2, 114--125. Google Scholar
Digital Library
- Comis, C. 2005. A high-speed inter-process communication architecture for FPGA-based hardware acceleration of molecular dynamics. M.S. thesis, University of Toronto.Google Scholar
- Cray Inc. 2005. CRAY XD1 FPGA Development. CRAY Inc. pp. 9-11, pp. 63-66.Google Scholar
- Cray, Inc. 2009a. http://www.cray.com/ {Accessed 3/09}.Google Scholar
- Cray, Inc. 2009b. Cray XT5h Supercomputer Brochure. http://www.cray.com/Assets/PDF/products/xt/CrayXT5hBrochure.pdf {Accessed 3/09}.Google Scholar
- DRC Computer. 2009. http://www.drccomputer.com/ {Accessed 3/09}.Google Scholar
- Ebcioglu, K., Saraswat, V., and Sarkar, V. 2005. X10: an Experimental Language for High Productivity Programming of Scalable Systems. In Proceedings of the 2nd International Workshop on Productivity and Performance in High-End Computing (P-PHEC).Google Scholar
- Graham, R. L., Woodall, T. S., and Squyres, J. M. 2005. Open MPI: A flexible high performance MPI. In Proceedings of the 6th Annual International Conference on Parallel Processing and Applied Mathematics. Google Scholar
Digital Library
- Gropp, W., Lusk, E., Doss, N., and Skjellum, A. 1996. A high-performance, portable implementation of the MPI message passing interface standard. Parall. Comput. 22, 6, 789--828. Google Scholar
Digital Library
- HyperTransport Consortium. 2009. http://www.hypertransport.org {Accessed 3/09}.Google Scholar
- Impulse Accelerated Technologies, Inc. 2009. http://www.impulsec.com {Accessed 3/09}.Google Scholar
- Intel. 2009a. Intel Quick Path Architecture (White Paper). http://www.intel.com/technology/quickpath/whitepaper.pdf {Accessed 3/09}.Google Scholar
- Intel. 2009b. Intel QuickAssist Technology AAL (White Paper). http://download.intel.com/technology/platforms/quickassist/quickassist_aal_whitepaper.pdf {Accessed 3/09}.Google Scholar
- Intel Corp. 2009. http://www.intel.com/ {Accessed 3/09}.Google Scholar
- Jerraya, A. and Wolf, W. 2004. Multiprocessor Systems-on-Chip. Morgan-Kaufmannn.Google Scholar
- Mentor Graphics. 2009. http://www.mentor.com/ {Accessed 3/09}.Google Scholar
- Mitrionics, Inc. 2009. http://www.mitrionics.com {Accessed 3/09}.Google Scholar
- Moore, N., Conti, A., Leeser, M., and King, L. 2007. Vforce: An extensible framework for reconfigurable supercomputing. Computer 40, 3, 39--49. Google Scholar
Digital Library
- MPI Parallel Environment. 2008. http://www-unix.mcs.anl.gov/perfvis/software/MPE/index.html.Google Scholar
- Nallatech, Inc. 2009. http://www.nallatech.com/ {Accessed 3/09}.Google Scholar
- Nunes, D., Saldaña, M., and Chow, P. 2008. A profiler for a heterogeneous multi-core multi-FPGA system. In Proceedings of the IEEE International Conference on Field-Programmable Technology.Google Scholar
- OpenFPGA 2009a. OpenFPGA. http://www.openfpga.org/ {Accessed 3/09}.Google Scholar
- OpenFPGA 2009b. OpenFPGA GenAPI version 0.4 Draft for Comment. http://www.openfpga.org/pages/Standards.aspx {Accessed 3/09}.Google Scholar
- Patel, A., Madill, C. A., Saldana, M., Comis, C., Pomes, R., and Chow, P. 2006. A scalable FPGA-based multiprocessor. In Proceedings of the 14th Annual IEEE Symposium on Field-Programmable Custom Computing Machines (FCCM’06). 111--120. Google Scholar
Digital Library
- Project Fortress Community. 2009. http://projectfortress.sun.com/Projects/Community/ {Accessed 3/09}.Google Scholar
- Saldaña, M. 2006. A parallel programming model for a multi-FPGA multiprocessor machine. M.S. dissertation, University of Toronto.Google Scholar
- Saldaña, M. and Chow, P. 2006. TMD-MPI: An MPI implementation for multiple processors across multiple FPGAs. In Proceedings of the IEEE 16th International Conference on Field-Programmable Logic and Applications.Google Scholar
- Saldaña, M., Nunes, D., Ramalho, E., and Chow, P. 2006. Configuration and programming of heterogeneous multiprocessors on a multi-FPGA system using TMD-MPI. In Proceedings of the IEEE 3rd International Conference on Reconfigurable Computing and FPGAs.Google Scholar
- Saldaña, M., Ramalho, E., and Chow, P. 2008. A message-passing hardware/software co-simulation environment to aid in reconfigurable computing design using TMD-MPI. In Proceeding of the IEEE International Conference on Reconfigurable Computing and FPGAs (ReConFig’08). Google Scholar
Digital Library
- SGI. 2008a. Reconfigurable Application-Specific Computing User’s Guide. SGI. 9--12; 223--244.Google Scholar
- SGI. 2008b. SGI RASC RC100 Blade. http://www.sgi.com/pdfs/3920.pdf {Accessed 3/09}.Google Scholar
- SGI. 2009. http://www.sgi.com/ {Accessed 3/09}.Google Scholar
- SRC Computers, Inc. 2009. http://www.srccomp.com/ {Accessed 3/09}.Google Scholar
- Synfora, Inc. 2009. http://www.synfora.com {Accessed 3/09}.Google Scholar
- The MPI Forum. 1993. MPI: A Message Passing Interface. In Proceedings of the ACM/IEEE Conference on Supercomputing (Supercomputing’93). ACM Press, New York, 878--883. Google Scholar
Digital Library
- Wang, D. 2008. An automated flow to generate hardware computing nodes from C for an FPGA-based MPI computing network. Engineering science undergraduate dissertation, University of Toronto.Google Scholar
- Xtreme Data Inc. 2009. http://www.xtremedatainc.com/ {Accessed 3/09}.Google Scholar
- Zaki, O., Lusk, E., Gropp, W., and Swider, D. 1999. Toward scalable performance visualization with jumpshot. Int. J. High Perform. Comput. Appl. 13, 3, 277--288. Google Scholar
Digital Library
Index Terms
MPI as a Programming Model for High-Performance Reconfigurable Computers
Recommendations
A Message-Passing Hardware/Software Co-simulation Environment to Aid in Reconfigurable Computing Design Using TMD-MPI
RECONFIG '08: Proceedings of the 2008 International Conference on Reconfigurable Computing and FPGAsHigh-performance reconfigurable computers (HPRC) provide a mix of standard processors and FPGAs to collectively accelerate applications. This introduces new design challenges, such as the need for portable programming models across HPRCs, and system-...
A reconfigurable HW/SW platform for computation intensive high-resolution real-time digital film applications
DATE '06: Proceedings of the conference on Design, automation and test in Europe: ProceedingsThis paper presents a multi-board, multi-FPGA hardware/software architecture, for computation intensive, high resolution (2048x2048 pixels), real-time (24 frames per second) digital film processing. It is based on Xilinx Virtex-II Pro FPGAs, large SDRAM ...
Design and implementation of a plesiochronous multi-core 4x4 network-on-chip FPGA platform with MPI HAL support
FPGAworld '09: Proceedings of the 6th FPGAworld ConferenceThe Multi-Core NoC is a 4 by 4 Mesh NoC targeted for Altera FPGAs. It implements a deflective routing policy and is used to connect sixteen NIOS II processors. Each NIOS II is connected to the NoC via an address-mapped Resource Network Interface.
The ...






Comments