SBIR Phase I: Reliable Distributed-Object Computing with Fault-Tolerant Cooperative Intelligent Agents

SBIR 第一阶段：具有容错协作智能代理的可靠分布式对象计算

• 批准号: 9861270
• 负责人: Jeffrey Yalowitz
• 金额: $ 9.99万
• 依托单位: SENTAR, INC
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1999
• 资助国家: 美国
• 起止时间: 1999-01-01 至 1999-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9861270&HistoricalAwards=false
• 关键词: SBIR; Phase; Reliable; Distributed; Object

9861270 This Small Business Innovation Research Phase I project builds on new cooperative intelligent agent technology and proven fault tolerance techniques to improve greatly the reliability of distributed object-computing environments such as the Common Object Request Brokering Architecture (CORBA). The movement toward Distributed Object Request Management (DOM) systems such as CORBA recognizes the need to extend object-oriented computing dynamically into the network. However, these architectures are vulnerable because of the variability of networks and the complexity of distributed interfaces, and they do not incorporate fault tolerance features. Fault tolerance is essential for DOM to he usable in many distributed systems with high value or safety implications. The goal is to make CORBA fault-tolerant by employing fault-tolerant cooperative intelligent agent developed by Sentar as, adjuncts to the distributed client applications, object request brokers, and object repositories that comprise CORBA implementations. The project will investigate fault tolerance methods for application and abstract levels of interaction that do not impact the CORBA standard or burden non-critical applications. The objectives are to adapt agent and fault tolerance software components for the CORBA environment, prototype critical aspects of the fault CORBA support agent network, and evaluate reliability under various feasible fault conditions. A vast commercial market has developed in recent years for distributed object computing technology and products. Of the market, a significant segment requires highly reliable software for safety and asset protection. If the proposed research meets its goal, users in this critical software segment will enjoy the cost savings and operational benefits CORBA provides to others.

9861270 这个小型企业创新研究第一阶段项目建立在新的协作智能代理技术和经过验证的容错技术的基础上，可大大提高分布式对象计算环境（例如公共对象请求代理架构（CORBA））的可靠性。诸如 CORBA 之类的分布式对象请求管理 (DOM) 系统的发展认识到需要将面向对象的计算动态扩展到网络中。然而，由于网络的可变性和分布式接口的复杂性，这些架构很容易受到攻击，并且它们不包含容错功能。容错对于 DOM 在许多具有高价值或安全隐患的分布式系统中使用至关重要。 目标是通过使用 Sentar 开发的容错协作智能代理来使 CORBA 具有容错能力，作为组成 CORBA 实现的分布式客户端应用程序、对象请求代理和对象存储库的附件。该项目将研究应用程序的容错方法和交互的抽象级别，这些方法不会影响 CORBA 标准或给非关键应用程序带来负担。 目标是使代理和容错软件组件适应 CORBA 环境，对故障 CORBA 支持代理网络的关键方面进行原型设计，并评估各种可行故障条件下的可靠性。 近年来，分布式对象计算技术和产品已经形成了广阔的商业市场。在市场中，很大一部分需要高度可靠的软件来实现安全和资产保护。如果拟议的研究达到其目标，这个关键软件领域的用户将享受 CORBA 为其他人提供的成本节约和运营优势。