Annual Report for Contract Number N00014-88-K-0641
Abstract
The primary goal of the Avalon Project is to create a set of linguistic constructs designed to give programmers explicit control over transaction-based processing of atomic objects for fault-tolerant applications. These constructs have been implemented as extensions to C++ and Common Lisp. The constructs include new encapsulation and abstraction mechanisms, as well as support for concurrency and recovery. The decision to extend an existing language rather than to invent a new one was based on pragmatic considerations. We felt we could focus more effectively on the new and interesting issues of reliability and concurrency if we did not have to redesign or reimplement basic language features, and we felt that building on top of a widely-used and widely-available language would facilitate the use of Avalon outside our own research group.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 30, 1990
- Accession Number
- ADA248439
Entities
People
- Nico Habermann
Organizations
- Carnegie Mellon University