DMCA
Experiences with the amoeba distributed operating system (1990)
Cached
Download Links
| Venue: | Communications of the ACM |
| Citations: | 228 - 20 self |
BibTeX
@ARTICLE{Tanenbaum90experienceswith,
author = {Andrew S. Tanenbaum and Robbert Van Renesse and Hans Van Staveren and Gregory J. Sharp and Sape J. Mullender and Jack Jansen and Guido Van Rossum},
title = {Experiences with the amoeba distributed operating system},
journal = {Communications of the ACM},
year = {1990},
volume = {33},
pages = {46--63}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
The Amoeba distributed operating system has been in development and use for over eight years now. In this paper we describe the present system and our experience with it—what we did right, but also what we did wrong. Among the things done right were basing the system on objects, using a single uniform mechanism (capabilities) for naming and protecting them in a location independent way, and designing a completely new, and very fast file system. Among the things done wrong were having threads not be pre-emptable, initially building our own homebrew window system, and not having a multicast facility at the outset.
