The Operational Analysis of Queueing Network Models
ACM Computing Surveys (CSUR)
Approximate Methods for Analyzing Queueing Network Models of Computing Systems
ACM Computing Surveys (CSUR)
High level programming for distributed computing
Communications of the ACM
Communicating sequential processes
Communications of the ACM
Distributed processes: a concurrent programming concept
Communications of the ACM
Ethernet: distributed packet switching for local computer networks
Communications of the ACM
TASK forces: Distributed software for solving problems of substantial size
ICSE '79 Proceedings of the 4th international conference on Software engineering
On the performance of decentralized software
PERFORMANCE '80 Proceedings of the 1980 international symposium on Computer performance modelling, measurement and evaluation
PERFORMANCE '80 Proceedings of the 1980 international symposium on Computer performance modelling, measurement and evaluation
Primitives for distributed computing
SOSP '79 Proceedings of the seventh ACM symposium on Operating systems principles
The behavior of Ethernet-like computer communications networks
SOSP '79 Proceedings of the seventh ACM symposium on Operating systems principles
| Hi-index | 0.00 |
The dominant features of software for distributed computer systems are communication between processes, potential parallel execution of processes, and non-determinism. This paper examines how a software architecture can be expected to use a distributed computer system based on a contention bus. Processes tend to be fine-grained and transmit short messages. An operating system kernel that implements process interactions introduces an additional source of short messages used to coordinate process interactions. Existing analyses of contention bus performance show that traffic patterns will be unable to utilize more than a fraction of the potential contention bus bandwidth. This paper suggests some possible ways to extend performance models to more accurately reflect the features of software for distributed computer systems. It also suggests an approach to improving channel efficiency for implementations of distributed software.