中圖分類號(hào)： TN919.5；TP393 文獻(xiàn)標(biāo)識(shí)碼： A DOI：10.16157/j.issn.0258-7998.200811 中文引用格式： 孫日明,，胡先浪. 一種面向智能網(wǎng)絡(luò)系統(tǒng)的自主計(jì)算能力分析方法[J].電子技術(shù)應(yīng)用,，2021，47(9)：59-63,，68. 英文引用格式： Sun Riming,，Hu Xianlang. An analysis method of autonomic computing capability for intelligent network system[J]. Application of Electronic Technique，2021,，47(9)：59-63,，68.
An analysis method of autonomic computing capability for intelligent network system
Sun Riming，Hu Xianlang
Jiangsu Automation Research Instisute,，Lianyungang 222061,，China
Abstract： At present, all kinds of intelligent network systems have been widely used, but due to the large number of nodes and complex external environment, its self-management has great challenges. The autonomous computing system(ACS) has the ability to manage itself according to the strategy and goal, and has broad application prospects in the complex intelligent network system. However, the current evaluation method of autonomic computing lacks accurate quantification to evaluate the self-management level of ACS. This paper firstly proposes an evaluation model of autonomic computing based on PEPA(performance evaluation process algbra). Then, according to the core idea of autonomic computing(less or no intervention),a self-management evaluation index is proposed. In addition, in order to avoid the state space explosion of traditional Markov chain caused by the huge scale of ACS, ODEs(ordinary differential equations) are generated from PEPA model by using continuous state space approximation method. The experimental results show that improving the detection success rate and self-* transition rate is of great significance to improve autonomic computing. This work provides an evaluation method for autonomic computing, which can automatically measure the self-management ability.
Key words : autonomic computing；PEPA,；continuous state space approximation
