The Community for Technology Leaders
RSS Icon
Subscribe
Issue No.01 - January-June (2013 vol.12)
pp: 17-20
ABSTRACT
RTL design complexity discouraged adoption of reconfigurable logic in general purpose systems, impeding opportunities for performance and energy improvements. Recent improvements to HLS compilers simplify RTL design and are easing this barrier. A new challenge will emerge: managing reconfigurable resources between multiple applications with custom hardware designs. In this paper, we propose a method to "shrink-fit" accelerators within widely varying fabric budgets. Shrink-fit automatically shrinks existing accelerator designs within small fabric budgets and grows designs to increase performance when larger budgets are available. Our method takes advantage of current accelerator design techniques and introduces a novel architectural approach based on fine-grained virtualization. We evaluate shrink-fit using a synthesized implementation of an IDCT for decoding JPEGs and show the IDCT accelerator can shrink by a factor of 16x with minimal performance and area overheads. Using shrink-fit, application designers can achieve the benefits of hardware acceleration with single RTL designs on FPGAs large and small.
INDEX TERMS
Accelerators, Field programmable gate arrays, Program processors, Computer applications, Decoding, Runtime,Special-Purpose and Application-Based Systems, Heterogeneous (hybrid) systems, Reconfigurable hardware
CITATION
M. Lyons, Gu-Yeon Wei, D. Brooks, "Shrink-Fit: A Framework for Flexible Accelerator Sizing", IEEE Computer Architecture Letters, vol.12, no. 1, pp. 17-20, January-June 2013, doi:10.1109/L-CA.2012.7
REFERENCES
1. G. Stitt,“Are Field-Programmable Gate Arrays Ready for the Mainstream?” IEEE Micro, vol. 31, no. 6, Nov. 2011.
2. G. Brebner,“A virtual hardware operating system for the Xilinx XC6200,” FPL, pp. 327-336, 1996.
3. J. Kelm and S. Lumetta,“HybridOS: runtime support for reconfigurable accelerators,” FPGA, pp. 212-221, 2008.
4. M. J. Lyons et al., “The Accelerator Store: A Shared Memory Framework For Accelerator-Based Systems,” ACM TACO, vol. 8, no. 4, pp. 1-22, Jan. 2012.
5. D. M. Harris and S. L. Harris,“Digital Design and Computer Architecture,” Morgan Kaufmann, pp. 167-168, 187, 233, 2007.
6. 192 AES. opencores.org/project, systemcaes
7. Reed Solomon. opencores.org/project, reed solomon decoder
8. Elliptic Curve Group. opencores.org/project, ecg
9. W. Thies and M. Karczmarek,“StreamIt: A language for streaming applications,” Compiler Construction, Jan. 2002.
10. M. Dales,“Managing a Reconfigurable Processor in a General Purpose Workstation Environment,” DATE, vol. 1, Mar. 2003.
11. S. Kumar et al., “C-core: Using communication cores for high performance network services,” NCA, Jan. 2005.
12. J. R. Wernsing and G. Stitt,“Elastic computing: a framework for transparent, portable, and adaptive multi-core heterogeneous computing,” in LCTES, Apr. 2010.
13. J. Cong,K. Gururaj,, and G. Han,“Synthesis of reconfigurable high-performance multicore systems,” FPGA, Feb. 2009.
14. C. Huang and F. Vahid,“Transmuting coprocessors: dynamic loading of FPGA coprocessors,” in DAC, Jul. 2009.
15. V. Rana et al., “Minimization of the reconfiguration latency for the mapping of applications on FPGA-based systems,” in CODES+ISSS, Oct. 2009.
78 ms
(Ver 2.0)

Marketing Automation Platform Marketing Automation Tool