Performance-Energy Tradeoffs for Matrix Multiplication on FPGA-Based Mixed-Mode Chip Multiprocessors

Xiaofang Wang1 and Sotirios Ziavras2
1Villanova University, 2New Jersey Institute of Technology


Abstract

Chip multiprocessing has demonstrated to be a promising approach in microprocessor design. With ever-increasing concerns for energy consumption, performance-energy trade-offs are often necessary, especially in the design of real-time embedded systems. This paper presents our performance and energy study on an in-house developed FPGA-based mixed-mode chip multiprocessor, where the SIMD (Single-Instruction, Multiple-Data), MIMD (Multiple-Instruction, Multiple-Data) and M-SIMD (Multiple-SIMD) computing modes can exist simultaneously in one system. We propose performance-energy trade-off techniques based on the observation that SIMD and MIMD task executions involve substantially different amounts of computation and communication, which result in different time and energy behavior and provide us with opportunities to realize various performance-energy objectives. Generalized matrix-matrix multiplication (MMM) is employed as an example to illustrate our analysis. Experimental results on a Xilinx Virtex II XC2V6000-5 FPGA demonstrate the effectiveness of the proposed approach.