학술논문

Optimal loop-unrolling mechanisms and architectural extensions for an energy-efficient design of shared register files in MPSoCs
Document Type
Conference
Source
Innovative Architecture for Future Generation High-Performance Processors and Systems (IWIA'05) Future Generation High-Performance Processors and Systems Innovative Architecture for Future Generation High-Performance Processors and Systems, 2005. :7 pp. 2005
Subject
Communication, Networking and Broadcast Technologies
Computing and Processing
Signal Processing and Analysis
Energy efficiency
Registers
Hardware
Computer architecture
Energy consumption
VLIW
Embedded software
Software performance
Multimedia systems
Embedded computing
Language
ISSN
1537-3223
Abstract
In this paper, we introduce a new hardware/software approach to reduce the energy of the shared register file in upcoming embedded architectures with several VLIW processors. This paper includes a set of architectural extensions and special loop unrolling techniques for the compilers of MPSoC platforms. This complete hardware/software support enables reducing the energy consumed in the register file of MPSoC architectures up to a 60% without introducing performance penalties.