Support for Data Sharing in a Heterogeneous Distributed System.

越来越多的强大而廉价的计算机位于个人用户附近，这些计算机要么本地连接，要么远距离连接在一起。这些计算机是不同的，因为用户需要不同的功能；未来似乎不太可能像过去一样，在任何组织中立法使用相同的系统。与并行处理一起，异构性和分布性很可能成为下一个十年计算机系统的主要特征。分布式、异质系统的出现带来了一些新的问题。一个特别重要的问题是缺乏对跨异类网络以安全和方便的方式共享程序和数据的支持。该项目旨在通过提供独立于语言的对象存储库来纠正此问题。存储库可以由用不同编程语言编写的程序使用，并在不同的硬件和不同的操作系统下运行，以共享信息。新的存储库将独立于语言，将存储类型化、结构化的对象，将支持原子事务，将是高可靠和高可用的，将控制对其对象的访问，将执行良好，并且将是可伸缩的，因此它可以支持广泛的系统规模。

Increasingly, powerful and inexpensive computers are situated near individual users, and these computers are linked together either locally or across long distances. The computers are heterogeneous because users require different capabilities; legislating the use of identical systems in any organization seems as unlikely in the future as it has been in the past. Together with parallel processing, heterogenity and distribution are likely to be the dominant characteristics of computer systems during the next decade. The advent of distributed, heterogeneous systems has led to some new problems. One particularly important problem is the lack of support for sharing of programs and data in a safe and convenient manner across a heterogeneous network. This project is aimed at correcting this problem by providing a language-independent object repository. The repository can be used by programs written in different programming languages, and running on different hardware and under different operating systems, to share information. The new repository will be language independent, will store typed, structured objects, will support atomic transactions, will be both highly reliable and highly available, will control access to its objects, will perform well, and will be scalable, so that it can support a wide range of system sizes.