The instrumentation of multics
Communications of the ACM
Operating Systems: On overcoming high-priority paralysis in multiprogramming systems: a case history
Communications of the ACM
A study of resource utilization and performance evaluation of large-scalecomputer systems
A study of resource utilization and performance evaluation of large-scalecomputer systems
Mathematical models of multiprogrammed computer systems
Mathematical models of multiprogrammed computer systems
Performance monitors for multi-programming systems
SOSP '69 Proceedings of the second symposium on Operating systems principles
An experimental study of computer system performance
ACM '72 Proceedings of the ACM annual conference - Volume 2
SOSP '71 Proceedings of the third ACM symposium on Operating systems principles
Simulation in the evaluation of an executive system design
ANSS '73 Proceedings of the 1st symposium on Simulation of computer systems
Minimization of supervisor conflict for multiprocessor computer systems
ANSS '76 Proceedings of the 4th symposium on Simulation of computer systems
Experience with trace driven modeling
ANSS '76 Proceedings of the 4th symposium on Simulation of computer systems
Performance of an I/O channel with multiple paging drums
SIGME '73 Proceedings of the 1973 ACM SIGME symposium
ACM SIGOPS Operating Systems Review
An approach to the collection, representation and simulation of program execution behavior
ACM SIGSIM Simulation Digest
Aspects of a Dynamically Adaptive Operating System
IEEE Transactions on Computers
Validation of a trace-driven CDC 6400 simulation
AFIPS '72 (Spring) Proceedings of the May 16-18, 1972, spring joint computer conference
The interaction of multi-programming job scheduling and CPU scheduling
AFIPS '72 (Fall, part I) Proceedings of the December 5-7, 1972, fall joint computer conference, part I
Hi-index | 0.00 |
Microscopic level job stream data obtained in a production environment by an event driven software probe is used to drive a model of a multiprogramming computer system. The CPU scheduling algorithm of the model is systematically varied. This technique, called trace driven modeling, provides an accurate replica of a production environment for the testing of variations in the system. At the same time alterations in scheduling methods can be easily carried out in a controlled way with cause and effects relationships being isolated. The scheduling methods tested included the best possible and worst possible methods, the traditional methods of multiprogramming theory, round-robin, first-come-first-serve, etc., and dynamic predictors. The relative and absolute performance of these scheduling methods is given. It is concluded that a successful CPU scheduling method must be pre-emptive and must prevent a given job from holding the CPU for too long a period.