Using Neural Networks to Forecast Available System Resources: An Approach and Empirical Investigation

被引：8

作者：

Jia, Yun-Fei ^{[1
]}

Zhou, Zhi Quan ^{[2
]}

Xue, Ke-Xian ^{[3
]}

Zhao, Lei ^{[4
]}

Cai, Kai-Yuan ^{[5
]}

机构：

[1] Civil Aviat Univ China, Tianjin Key Lab Adv Signal Proc, Tianjin 300300, Peoples R China

[2] Univ Wollongong, Sch Comp & Informat Technol, Wollongong, NSW 2522, Australia

[3] Inst NBC Def PLA, Nanjing, Jiangsu, Peoples R China

[4] Beijing Inst Control Engn, Beijing 100080, Peoples R China

[5] Beijing Univ Aeronaut & Astronaut, Dept Automat Control, Beijing 100191, Peoples R China

来源：

INTERNATIONAL JOURNAL OF SOFTWARE ENGINEERING AND KNOWLEDGE ENGINEERING | 2015年 / 25卷 / 04期

基金：

澳大利亚研究理事会;

关键词：

Forecasting; neural networks; software aging; software reliability; system availability; system resources; system workload; SOFTWARE REJUVENATION; ASSURANCE; MODEL;

D O I：

10.1142/S0218194015500102

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Software aging refers to the phenomenon that software systems show progressive performance degradation or a sudden crash after longtime execution. It has been reported that this phenomenon is closely related to the exhaustion of system resources. This paper quantitatively studies available system resources under the real-world situation where workload changes dynamically over time. We propose a neural network approach to first investigate the relationship between available system resources and system workload and then to forecast future available system resources. Experimental results on data sets collected from real-world computer systems demonstrate that the proposed approach is effective.

引用

页码：781 / 802

页数：22

共 36 条

[1] Using machine learning for non-intrusive modeling and prediction of software aging [J].

Andrzejak, Artur ;

Silva, Luis .

2008 IEEE NETWORK OPERATIONS AND MANAGEMENT SYMPOSIUM, VOLS 1 AND 2, 2008, :25-+

[2]

Astrom K. J., 1996, COMPUTER CONTROLLED, V3rd

[3] Long-term wind speed and power forecasting using local recurrent neural network models [J].

Barbounis, TG ;

Theocharis, JB ;

Alexiadis, MC ;

Dokopoulos, PS .

IEEE TRANSACTIONS ON ENERGY CONVERSION, 2006, 21 (01) :273-284

[4] An efficient parameterization of dynamic neural networks for nonlinear system identification [J].

Becerra, VM ;

Garces, FR ;

Nasuto, SJ ;

Holderbaum, W .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 2005, 16 (04) :983-988

[5] SINGLE-LAYER NEURAL NETWORKS FOR LINEAR-SYSTEM IDENTIFICATION USING GRADIENT DESCENT TECHNIQUE [J].

BHAMA, S ;

SINGH, H .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1993, 4 (05) :884-888

[6] Workload-Based Software Rejuvenation in Cloud Systems [J].

Bruneo, Dario ;

Distefano, Salvatore ;

Longo, Francesco ;

Puliafito, Antonio ;

Scarpa, Marco .

IEEE TRANSACTIONS ON COMPUTERS, 2013, 62 (06) :1072-1085

[7] Software reliability experimentation and control [J].

Cai, Kai-Yuan .

JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY, 2006, 21 (05) :697-707

[8]

Chen XE, 2006, SOSE 2006: SECOND IEEE INTERNATIONAL SYMPOSIUM ON SERVICE-ORIENTED SYSTEM ENGINEERING, PROCEEDINGS, P34

[9]

Dohi T, 2000, 2000 PACIFIC RIM INTERNATIONAL SYMPOSIUM ON DEPENDABLE COMPUTING, PROCEEDINGS, P77, DOI 10.1109/PRDC.2000.897287

[10] Estimating software rejuvenation schedules in high-assurance systems [J].

Dohi, T ;

Gogeva-Popstojanova, K ;

Trivedi, K .

COMPUTER JOURNAL, 2001, 44 (06) :473-485

← 1 2 3 4 →