Energy metric for software systems

被引:11
作者
Chatzigeorgiou, A [1 ]
Stephanides, G [1 ]
机构
[1] Univ Macedonia, Dept Appl Informat, Thessaloniki 54006, Greece
关键词
software metrics; energy consumption; embedded software;
D O I
10.1023/A:1022142105380
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
Acknowledging the intense requirement for low power operation in most portable computing systems, this paper introduces the notion of energy efficient software design and proposes metrics, for evaluating software systems in terms of their energy consumption. Considering the sources of power consumption in every digital circuit, and the fact that power is primarily dependent on the executing software, appropriate energy measures are derived, which can be extracted from the flowgraph of a program. The proposed measures are computed by applying rules common to the existing hierarchical measures of other internal software attributes, and form the basis for the definition of a software energy metric. This metric can be used in order to determine the level of energy consumption of any software system more efficiently than existing assembly-parsing techniques, with only a limited penalty in accuracy. Application to different implementations of algorithms, drawn from matrix algebra and multimedia, demonstrates the efficiency of the proposed energy metric for comparison purposes, and as an indicator for quality improvement.
引用
收藏
页码:355 / 371
页数:17
相关论文
共 27 条
[1]  
[Anonymous], IMAGE VIDEO COMPRESS
[2]  
[Anonymous], P 36 ACM IEEE C DES
[3]  
[Anonymous], SOFTWARE ENG FORMAL
[4]   System-level power optimization: Techniques and tools [J].
Benini, L ;
De Micheli, G .
ACM TRANSACTIONS ON DESIGN AUTOMATION OF ELECTRONIC SYSTEMS, 2000, 5 (02) :115-192
[5]  
Brooks D, 2000, PROCEEDING OF THE 27TH INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE, P83, DOI [10.1145/342001.339657, 10.1109/ISCA.2000.854380]
[6]  
Catthoor Francky, 1998, CUSTOM MEMORY MANAGE
[7]  
Chandrakasan A.P., 1995, Low Power Digital CMOS Design
[8]   A memory management approach for efficient implementation of multimedia kernels on programmable architectures [J].
Dasigenis, M ;
Kroupis, N ;
Argyriou, A ;
Tatas, K ;
Soudris, D ;
Thanailakis, A ;
Zervas, N .
IEEE COMPUTER SOCIETY WORKSHOP ON VLSI 2001, PROCEEDINGS, 2001, :171-176
[9]  
FENTON N, 1996, SOFTWARE METRICS RIG
[10]   Power estimation of embedded systems: A hardware/software codesign approach [J].
Fornaciari, W ;
Gubian, P ;
Sciuto, D ;
Silvano, C .
IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS, 1998, 6 (02) :266-275