Numerical Methods for Scientists and Engineers
Numerical Methods for Scientists and Engineers
An approximate analysis of multiprocessor systems
SIGMETRICS '83 Proceedings of the 1983 ACM SIGMETRICS conference on Measurement and modeling of computer systems
Measurement and Analysis of HYPERchannel Networks
IEEE Transactions on Computers
Analysis of Memory Interference in Multiprocessors
IEEE Transactions on Computers
Bandwidth of Crossbar and Multiple-Bus Connections for Multiprocessors
IEEE Transactions on Computers
Demand Assignment Multiple Access Schemes in Broadcast Bus Local Area Networks
IEEE Transactions on Computers
The control data loosely coupled network lower level protocols
AFIPS '80 Proceedings of the May 19-22, 1980, national computer conference
Frequency-time controlled (FTC) networks for high speed communication
SIGCOMM '86 Proceedings of the ACM SIGCOMM conference on Communications architectures & protocols
Dynamic bandwidth allocation in a network
SIGCOMM '88 Symposium proceedings on Communications architectures and protocols
Approximate Analysis of Single and Multiple Ring Networks
IEEE Transactions on Computers
Broadcast-Efficient Protocols for Mobile Radio Networks
IEEE Transactions on Parallel and Distributed Systems
Energy-Efficient Routing in the Broadcast Communication Model
IEEE Transactions on Parallel and Distributed Systems
Hi-index | 15.00 |
The HYPERchannel communication network based on one to four trunks, or channels is considered. We develop closed queueing models with dependent servers which characterize the network performance as a function of the number of channels, the channel load, the number of stations and the packet length distribution. For analyzing the network behavior with constant packet length we introduce techniques for representing the actual asynchronous network operation by a tractable "sequential synchronous" model. Throughput and delay measures are in this way obtained for asynchronous systems operating with constant or exponential service times. The developed approach can also be used for obtaining tractable analyses of other asynchronous systems, such as multiprocessors employing multiple-bus interconnection. In the case of HYPERchannel networks, we apply the introduced models for analyzing multi- trunk network behavior and for predicting the number of channels needed to obtain a required packet-delay/trunk-utilization performance. In HYPERchannel networks the observed results are of special significance since they are most pronounced when the number of stations is small, the situation found in the majority of operational networks.