基于Java的Agents的安全问题(英文)

Security Issues for Java-Based Agents

在分布式系统中,随着计算机网络的普及,编程方式也发生了变化,即由远程调用方式转变为远距编程方式,这种变化可减少网络流量、提高系统性能.软件agents适合完成这项任务.使用agents能带来多种好处,且不利方面很少,这包括附加的安全问题.新的组件系统能改进Java套装软件系统的多个缺陷,使编程者能获得更多的对动态连接的控制权,实现Java-agent的自主安全.agent模型中出现的一般安全问题,可通过一种实用的方法加以解决.

With the popularization of computer networks,there has been a shift proposed in distributed system programming from the remote procedure call to the remote programming paradigm,to decrease network traffic and improve performance.Software agents could be used to accomplish this task.The use of agents has several advantages and a few disadvantages,including added security issues.In order to implement autonomous security of Java agent,the author presents a new module system for Java that improves upon many of the deficiencies of the Java package system and gives the programmer more control over dynamic linking.Next,the author develops a general agent model and discusses general security issues in that model.Finally,the author proposes a practical solution that addresses some of those security issues.