An analysis of some time-sharing techniques
Communications of the ACM
ECSS - an Extendable Computer System Simulator
Proceedings of the third conference on Applications of simulation
Simulation of the time-varying load on future remote-access immediate-response computer systems
Proceedings of the third conference on Applications of simulation
A Monte Carlo Process for Determining Response Times for Tactical Systems
Proceedings of the second conference on Applications of simulations
Proceedings of the second conference on Applications of simulations
Structured model development techniques
ANSS '74 Proceedings of the 2nd symposium on Simulation of computer systems
Computer system modelling: A test bed for new software technologies
WSC '74 Proceedings of the 7th conference on Winter simulation - Volume 2
Simulation of computer systems using automatically generated load descriptions
WSC '74 Proceedings of the 7th conference on Winter simulation - Volume 2
Use of performance analysis statistics in computer system simulation
WSC '71 Proceedings of the 5th conference on Winter simulation
DAS: An automated system to support design analysis
DAC '78 Proceedings of the 15th Design Automation Conference
DAS - an automated system to support design analysis
ICSE '78 Proceedings of the 3rd international conference on Software engineering
Hi-index | 0.00 |
The fundamental nature of computer systems is to serve the user. If we wish to evaluate the performance of a computer system, we must therefore seek to characterize its responsiveness to those actions initiated by the user. It is this responsibility that isolates accurate workload distributions as an imperative ingredient in computer systems evaluation, and it is this ingredient that the man/machine workload model addresses.In particular, the model presented here simulates human activity at a terminal in real-time or time-share computer system environments. The use of flowcharts to represent human decision making and a variable language definition capability to code flowchart actions are key to the approach. The techniques described stress automatic and manual verification as well as the ability to accurately represent workloads at the point where workloads originate, the human.With this representative approach, accurate workload generation is a natural outcome. In addition, experience has shown that the man/machine workload model serves to enhance computer system simulation credibility, provide a tradeoff tool for human factors and operations design "what-if" questions, and most importantly, to provide a validation methodology for front-end system design efforts.