ABSTRACT
Spatial Computing (SC) has been shown to be an energy-efficient model for implementing program kernels. In this paper we explore the feasibility of using SC for more than small kernels. To this end, we evaluate the performance and energy efficiency of entire applications on Tartan, a general-purpose architecture which integrates a reconfigurable fabric (RF) with a superscalar core. Our compiler automatically partitions and compiles an application into an instruction stream for the core and a configuration for the RF. We use a detailed simulator to capture both timing and energy numbers for all parts of the system.Our results indicate that a hierarchical RF architecture, designed around a scalable interconnect, is instrumental in harnessing the benefits of spatial computation. The interconnect uses static configuration and routing at the lower levels and a packet-switched, dynamically-routed network at the top level. Tartan is most energyefficient when almost all of the application is mapped to the RF, indicating the need for the RF to support most general-purpose programming constructs. Our initial investigation reveals that such a system can provide, on average, an order of magnitude improvement in energy-delay compared to an aggressive superscalar core on single-threaded workloads.
- M. Beck, R. Johnson, et al. From control flow to data flow. Journal of Parallel and Distributed Computing, 12:118--129, 1991. Google Scholar
Digital Library
- T. Bjerregaard and J. Sparsø. A scheduling discipline for latency and bandwidth guarantees in asynchronous Network-on-Chip. In International Symposium on Advanced Research in Asynchronous Circuits and Systems (ASYNC). IEEE, 2005. Google Scholar
Digital Library
- D. Brooks, V. Tiwari, et al. Wattch: a framework for architectural level power analysis and optimizations. In International Symposium on Computer Architecture (ISCA), pages 83--94. ACM Press, 2000. Google Scholar
Digital Library
- M. Budiu, P.V. Artigas, et al. Dataflow: A complement to superscalar. In IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), pages 177--186, March 20-22 2005. Google Scholar
Digital Library
- M. Budiu and S.C. Goldstein. Compiling application-specific hardware. In International Conference on Field Programmable Logic and Applications (FPL), pages 853--863, September 2002. Google Scholar
Digital Library
- M. Budiu and S.C. Goldstein. Pegasus: An efficient intermediate representation. Technical Report CMU-CS-02-107, Carnegie Mellon University, May 2002.Google Scholar
- M. Budiu, M. Mishra, et al. Peer-to-peer hardware-software interfaces for reconfigurable fabrics. In IEEE Symposium on Field-Programmable Custom Computing Machines (FCCM), pages 57--66, April 2002. Google Scholar
Digital Library
- M. Budiu, G. Venkataramani, et al. Spatial computation. In International Conference on Architectural Support for Programming Languages and Operating Systems, pages 14--26, October 2004. Google Scholar
Digital Library
- T. Callahan and J. Wawrzynek. Adapting software pipelining for reconfigurable computing. In Intl. Conf. on Compilers, Architecture, and Synthesis for Embedded Systems (CASES). ACM, 2000. Google Scholar
Digital Library
- L. Carter, B. Simon, et al. Path analysis and renaming for predicated instruction scheduling. International Journal of Parallel Programming, special issue, 28(6), 2000. Google Scholar
Digital Library
- T. Chelcea and S. Nowick. Robust interfaces for mixed-timing systems. In IEEE Transactions on Very Large Scale Integration (VLSI) Systems, volume 12-8, pages 857--873, 2004. Google Scholar
Digital Library
- R. Cytron, J. Ferrante, et al. Efficiently computing static single assignment form and the control dependence graph. ACM Transactions on Programming Languages and Systems (TOPLAS), 13(4):451--490, 1991. Google Scholar
Digital Library
- M.B. Gokhale, J.M. Stone, et al. Co-synthesis to a hybrid RISC/FPGA architecture. J. VLSI Signal Process. Syst., 24(2-3):165--180, 2000. Google Scholar
Digital Library
- S.C. Goldstein. The impact of the nanoscale on computing systems. In IEEE/ACM International Conference on Computer-aided design (ICCAD), 2005. Google Scholar
Digital Library
- S.C. Goldstein, H. Schmit, et al. PipeRench: a coprocessor for streaming multimedia acceleration. In International Symposium on Computer Architecture (ISCA), pages 28--39, May 1999. Google Scholar
Digital Library
- S. Hauck, S.M. Burns, et al. An FPGA for implementing asynchronous circuits. IEEE Design & Test of Computers, 11(3):60--69, 1994. Google Scholar
Digital Library
- J.R. Heath, P.J. Kuekes, et al. A defect-tolerant computer architecture: Opportunities for nanotechnology. Science, 280, 1998.Google Scholar
- N. Huot, H. Dubreuil, et al. FPGA architecture for multi-style asynchronous logic. In Design, Automation and Test in Europe (DATE), pages 32--33. IEEE Computer Society, 2005. Google Scholar
Digital Library
- Intel Corp. Intel Pentium M Datasheet, January 2006.Google Scholar
- J. Kao, S. Narendra, et al. Subthreshold leakage modeling and reduction techniques. In Proceedings of the 2002 IEEE/ACM International Conference on Computer Aided Design (ICCAD), pages 141--148, 2002. Google Scholar
Digital Library
- I. Kuon and J. Rose. Measuring the gap between FPGAs and ASICs. In Proceedings of the International Symposium on Field Programmable ate Arrays (FPGA'06), pages 21-30, February 2006. Google Scholar
Digital Library
- E. Larson, S. Chatterjee, et al. MASE: A novel architecture or detailed microarchitectural modeling. In IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), November 4-6 2001.Google Scholar
- C. Lee, M. Potkonjak, et al. MediaBench: a tool for evaluating and synthesizing multimedia and communications systems. In IEEE/ACM International Symposium on Microarchitecture (MICRO), pages 330--335, 1997. Google Scholar
Digital Library
- S.A. Mahlke, D.C. Lin, et al. Effective compiler support for predicated execution using the hyperblock. In International Symposium on Computer Architecture (ISCA), pages 45--54, Dec 1992. Google Scholar
Digital Library
- K. Mai, T. Paaske, et al. Smart memories: A modular reconfigurable architecture. In International Symposium on Computer Architecture (ISCA), June 2000. Google Scholar
Digital Library
- B.J. Nelson. Remote procedure call. Technical Report CSL-81-9, Xerox Palo Alto Research Center, 1981.Google Scholar
Digital Library
- R. Payne. Self-timed FPGA systems. In W. Moore and W. Luk, editors, International Conference on Field Programmable Logic and Applications (FPL), volume 975 of Lecture Notes in Computer Science, pages 21--35. Springer, 1995. Google Scholar
Digital Library
- R. Razdan and M.D. Smith. A High-Performance Microarchitecture with Hardware-Programmable Functional Units. In IEEE/ACM International Symposium on Microarchitecture (MICRO), pages 172--80. IEEE/ACM, November 1994. Google Scholar
Digital Library
- S. Rixner, W.J. Dally, et al. A bandwidth-efficient architecture for media processing. In IEEE/ACM International Symposium on Microarchitecture (MICRO), December 1998. Google Scholar
Digital Library
- K. Sankaralingam, R. Nagarajan, et al. Exploiting ILP, TLP, and DLP with the polymorphous TRIPS architecture. In International Symposium on Computer Architecture (ISCA), pages 422--433. ACM Press, 2003. Google Scholar
Digital Library
- H. Schmit, D. Whelihan, et al. Piperench: A virtualized programmable datapath in 0.18 micron technology. In IEEE Custom Integrated Circuits Conference, pages 63--66, 2002.Google Scholar
Cross Ref
- L. Shang and N. Jha. High-level power modeling of CPLDs and FPGAs. In International Conference on Computer Design (ICCD), pages 46--51, September 2001. Google Scholar
Digital Library
- L. Shang, A.S. Kaviani, et al. Dynamic power consumption in Virtex-II FPGA family. In ACM/SIGDA International Symposium on Field Programmable Gate Arrays (FPGA), pages 157--164. ACM Press, 2002. Google Scholar
Digital Library
- Standard Performance Evaluation Corp. SPEC INT 95 Benchmark Suite, 1995.Google Scholar
- Standard Performance Evaluation Corp. SPEC INT 2000 Benchmark Suite, 2000.Google Scholar
- I. Sutherland. Micropipelines: Turing award lecture. Communications of the ACM, 32 (6):720--738, June 1989. Google Scholar
Digital Library
- S. Swanson, K. Michelson, et al. Wavescalar. In IEEE/ACM International Symposium on Microarchitecture (MICRO), pages 291--302, December 2003. Google Scholar
Digital Library
- M.B. Taylor, W. Lee, et al. Evaluation of the RAW microprocessor: An exposed-wire-delay architecture for ILP and streams. In International Symposium on Computer Architecture (ISCA), pages 2--13. IEEE Computer Society, 2004. Google Scholar
Digital Library
- J. Teifel and R. Manohar. An asynchronous dataflow FPGA architecture. IEEE Trans. Computers, 53(11):1376--1392, 2004. Google Scholar
Digital Library
- W. Tsu, K. Macy, et al. HSRA: high-speed, hierarchical synchronous reconfigurable array. In ACM/SIGDA International Symposium on Field Programmable Gate Arrays (FPGA), pages 125--134. ACM Press, 1999. Google Scholar
Digital Library
- G. Venkataramani, T. Bjerregaard, et al. SOMA: a tool for synthesizing and optimizing memory accesses in ASICs. In International Symposium on Hardware/Software Codesign and System Synthesis (CODES+ISSS), pages 231--236. ACM Press, 2005. Google Scholar
Digital Library
- G. Venkataramani, M. Budiu, et al. C to asynchronous dataflow circuits: An end-to-end toolflow. In International Workshop on Logic Synthesis, June 2004.Google Scholar
- G. Venkataramani, T. Chelcea, et al. HLS support for unconstrained memory accesses. In International Workshop on Logic Syntheis, June 2005.Google Scholar
- M. Wazlowski, L. Agarwal, et al. PRISM-II compiler and architecture. In IEEE Symposium on Field-Programmable Custom Computing Machines (FCCM), pages 9--16, Apr 1993.Google Scholar
Cross Ref
- C. Wong, A. Martin, et al. An architecture for asynchronous FPGAs. In Proceedings of Field Programmable Technology (FPT), pages 170--177, 2003.Google Scholar
Cross Ref
- A.Z. Ye, A. Moshovos, et al. CHIMAERA: A high-performance architecture with a tightly-coupled reconfigurable unit. In International Symposium on Computer Architecture (ISCA), ACM Computer Architecture News. ACM Press, 2000. Google Scholar
Digital Library
Index Terms
Tartan: evaluating spatial computation for whole program execution
Recommendations
Spatial computation
ASPLOS XI: Proceedings of the 11th international conference on Architectural support for programming languages and operating systemsThis paper describes a computer architecture, Spatial Computation (SC), which is based on the translation of high-level language programs directly into hardware structures. SC program implementations are completely distributed, with no centralized ...
Tartan: evaluating spatial computation for whole program execution
Proceedings of the 2006 ASPLOS ConferenceSpatial Computing (SC) has been shown to be an energy-efficient model for implementing program kernels. In this paper we explore the feasibility of using SC for more than small kernels. To this end, we evaluate the performance and energy efficiency of ...
Tartan: evaluating spatial computation for whole program execution
Proceedings of the 2006 ASPLOS ConferenceSpatial Computing (SC) has been shown to be an energy-efficient model for implementing program kernels. In this paper we explore the feasibility of using SC for more than small kernels. To this end, we evaluate the performance and energy efficiency of ...








Comments