Performance gains from partitioning embedded applications in Processor-FPGA socs

Authors:
Michalis D. Galanis;Gregory Dimitroulakos;Costas E. Goutis
Affiliations:
VLSI Design Laboratory, ECE Department, University of Patras;VLSI Design Laboratory, ECE Department, University of Patras;VLSI Design Laboratory, ECE Department, University of Patras
Venue:
PATMOS'05 Proceedings of the 15th international conference on Integrated Circuit and System Design: power and Timing Modeling, Optimization and Simulation
Year:
2005

Citing 7
Cited 0

A low power hardware/software partitioning approach for core-based embedded systems

Proceedings of the 36th annual ACM/IEEE Design Automation Conference
A C compiler for a processor with a reconfigurable functional unit

FPGA '00 Proceedings of the 2000 ACM/SIGDA eighth international symposium on Field programmable gate arrays
The Garp Architecture and C Compiler

Computer
Energy Advantages of Microprocessor Platforms with On-Chip Configurable Logic

IEEE Design & Test
A C to Hardware/Software Compiler

FCCM '00 Proceedings of the 2000 IEEE Symposium on Field-Programmable Custom Computing Machines
The chimaera reconfigurable functional unit

IEEE Transactions on Very Large Scale Integration (VLSI) Systems
SpecSyn: an environment supporting the specify-explore-refine paradigm for hardware/software system design

IEEE Transactions on Very Large Scale Integration (VLSI) Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we propose a hardware/software partitioning method for improving performance in single-chip embedded systems comprised by processor and Field Programmable Gate Array reconfigurable logic. Speedups are achieved by executing critical software parts on the reconfigurable logic. A generic hybrid System-on-Chip platform, which can model existing processor-FPGA systems, is considered. The partitioning flow utilizes an automated analysis procedure at the basic-block level for detecting kernels in software. Three different instances of the considered generic platform and two sets of benchmarks are used in the experiments. For the systems composed by 32-bit processors the speedup of five applications ranges from 1.3 to 3.7 relative to an all software solution. For an 8-bit platform, the speedups of eight DSP algorithms are considerably greater, since they range from 3.2 to 68.4.