Journal article
Performance modelling of message-based multiprocessors under heavy traffic
Computer Systems Science and Engineering, Vol.7(3), pp.190-198
01 Jul 1992
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
This paper examines a model of a processing element in a message-passing multiprocessor system under heavy message traffic. Messages of random length arrive continuously, with no time between the end of one transmission and the beginning of the next one, at a queue in the input port of the processing element. Messages are processed locally and depart as new messages. Since there is no unused time between successive message arrivals, the time between successive arrivals is directly proportional to the message length. If the processing time is also related to the message length, then the processing element can be modelled as a single server queue with inter-dependent service and arrival times. This paper examines the case where processing time is also directly dependent on message length. Then, the ratio of processing time to interarrival time is constant, and, when less than 1, results is a stable system, which is studied using an integral relationship of the system time probability density function. This paper develops solutions for a number of inter-arrival time probability densities. The queue state probabilities just before an arrival are calculated from the system time densities. Theoretical results are compared with extensive simulation results.
Metrics
1 Record Views
Details
- Title
- Performance modelling of message-based multiprocessors under heavy traffic
- Creators
- Constantine Katsinis
- Publication Details
- Computer Systems Science and Engineering, Vol.7(3), pp.190-198
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- Computer Science (Computing)
- Identifiers
- 991020546598304721
UN Sustainable Development Goals (SDGs)
This output has contributed to the advancement of the following goals:
InCites Highlights
These are selected metrics from InCites Benchmarking & Analytics tool, related to this output
- Web of Science research areas
- Computer Science, Hardware & Architecture
- Computer Science, Theory & Methods