Mobile Agent Model for Flexible Distributed Systems.

在传统的客户机-服务器应用程序中，应用程序和对象分别分布到客户机和服务器中。另一方面，在对等（P2P）系统中，每台被称为服务器的计算机都是对等的，可以拥有程序和对象。此外，大量和各种类型的计算机在网络中互连，从高性能、高可靠性的计算机（如数据库服务器）到不太可靠的个人计算机和移动的站。在P2P系统中，如何在数据库上开发和执行可靠的应用程序是一个关键问题。在数据库应用程序中，事务操作对象以满足ACID（原子性、一致性、隔离性和持久性）属性。在传统的数据库系统中，对象被锁定或事务被加时间戳以序列化多个事务。如果操作多台计算机中的对象，则两阶段提交（2 PC）原型必须支持多台计算机的原子操作。 关于我们 在这项研究中，我们讨论了一种新型的代理模型，事务代理模型，以操纵对象在多台计算机上的类型的承诺约束。事务代理被定义为一个移动的代理，它在计算机周围移动，并在一台或多台计算机上本地操作具有承诺约束类型的对象。此外，如果代理以冲突的方式操纵同一对象，则代理与另一代理进行协商。通过协商，每个代理自主地决定代理是否持有或释放对象。如果代理要移动的计算机发生故障，则事务代理自主地找到另一个目的地计算机。在本研究中，我们讨论了如何在Java Aglets和Oracle数据库系统中实现事务代理。我们展示了应用程序的移动的基于代理的交易系统的评估。如果Aglet类是先验加载的，那么事务代理可以比传统的客户端-服务器模型更快地操作对象服务器。少

In traditional client-server applications, application programs and objects are distributed into clients and servers, respectively. On the other hand, every computer named servant is peer which can have programs and objects in peer-to-peer (P2P) systems. In addition, huge number and various types of computers are interconnected in networks, ranging from high-performance, high-reliable computers like database servers to less-reliable personal computers and mobile stations. It is critical to discuss how to develop and execute reliable applications on databases in P2P systems. In database applications, transactions manipulate objects so as to satisfy ACID (atomicity, consistency, isolation, and durability) properties. In traditional database systems, objects are locked or transactions are timestamped to serialize multiple transactions. If objects in multiple computers are manipulated, the atomic manipulation of multiple computers has to be supported in the two-phase commitment (2PC) proto … More col.In this research, we discussed a novel type of agent model, transactional agent model to manipulate objects in multiple computers with types of commitment constraints. A transactional agent is defined to be a mobile agent which moves around computers and locally manipulates objects in one or more than one computer with type of commitment constraint. In addition, an agent negotiates with another agent if the agents manipulate a same object in a conflicting manner. Through the negotiation, each agent autonomously makes a decision on whether the agent holds or releases the objects. A transactional agent autonomously finds another destination computer if a computer where the agent to move is faulty. In this research, we discussed how to implement transactional agents in Java Aglets and Oracle database systems. We showed the evaluation of the mobile agent-based transaction systems for applications. If Aglets classes are a priori loaded, the transactional agents can manipulate object servers more shortly than the traditional client-server model. Less