Topics in Concurrent Programming

该项目涉及设计和使用中的几个相关主题， 并发程序设计语言 一、最新实施的 SR分布式编程语言的一个新版本正在完善中， 它可以被感兴趣的各方获取和使用。 二是各种 目前正在研究对该语言的扩展及其实施。 具体的主题包括提供将新程序链接到 执行SR程序，为数据提供有效支持 抽象，开发独立的实现，并实现 一个强大的宏处理器，以方便新的实验 语言机制。 第三，容错编程的各个方面 正在调查 这项工作包括制定、实施和 实验语言机制来处理异步 处理器故障的性质;并设计分析技术， 确定容错程序的性能。 四、工作 继续在MLP系统上构建分布式混合- 语言程序。 这里的重点是向 通过将SR作为支持的语言之一，开发 使用MLP的分布式应用程序，并扩展系统以包括 更多的顺序语言 第五，正在研究如何 扩展SR以提供对并行编程的支持;特定主题 包括提供全球性、共享的业务，并提供机制， 用于组通信。 正在探讨的最后一个主题是 发展的范例，系统地解决分布式和 并行编程问题。 预计这项研究将带来重大改进 在我们对构建并发 系统.

This project addresses several related topics in the design and use of concurrent program languages. First, the implementation of the latest version of the SR distributed programming language is being polished so it can be acquired and used by interested parties. Second, various extensions to the language and its implementation are being examined. Specific topics include providing facilities to link new programs into an executing SR program, providing efficient support for data abstraction, developing a stand-alone implementation, and implementing a powerful macro-processor to facilitate experimentation with new language mechanisms. Third, aspects of fault-tolerant programming are being investigated. This work includes developing, implementing, and experimenting with language mechanisms to handle the asynchronous nature of processor failures; and devising analytical techniques for determining the performance of fault-tolerant programs. Fourth, work is continuing on the MLP system for constructing distributed mixed- language programs. The focus here is on adding concurrency to the system by including SR as one of the supported languages, developing distributed applications using MLP, and extending the system to include additional sequential languages. Fifth, ways are being examined to extend SR to provide support for parallel programming; specific topics include providing global, shared operations and providing mechanisms for group communication. The final topic being explored is the development of paradigms for systematically solving distributed and parallel programming problems. It is expected that this research will lead to significant improvements in our understanding of language support issues for building concurrent systems.