Filters








358 Hits in 4.2 sec

Trace-driven modeling and analysis of CPU scheduling in a multiprogramming system

Stephen Sherman, Forest Baskett, J. C. Browne
1972 Communications of the ACM  
This technique, called trace-driven modeling, provides an accurate replica of a production environment for the testing of variations in the system.  ...  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 Trace-Driven Model The data used in this study was collected on a CDC 6600 running under the UT-1 operating system by the use of an event-driven software probe [8] .  ... 
doi:10.1145/361598.361626 fatcat:rredrgqxofb75jxei7uqyq737q

Trace-driven system modeling

P. S. Cheng
1969 IBM Systems Journal  
Simulation model The trace-driven approach was used in an experiment directed toward the determination of gross system performance as various multiprogramming and multiprocessing system configuration changes  ...  Once the system configuration has been specified and resources allocated, the primary function of the model is to time these CPU-input/output events in accordance with the scheduling and dispatching procedures  ... 
doi:10.1147/sj.84.0280 fatcat:lj4i4t3k4nhfphnsjhub5sd3oa

Page 25 of Computer Performance Vol. 5, Issue 1 [page]

1984 Computer Performance  
The simulation system is written in SIMSCRIPT”, and it can be driven either by the built-in stochastic semiMarkov models or by traces of real programs.  ...  Example test result Performance measure System driven by traces of real programs Relative error System driven by semiMarkov models U(CPU) E, (CPU) 0g (CPU) U(1/O) E,(1/O) 0q(1/O) 0.782 26 1.195 62 1.16005  ... 

The interaction of multi-programming job scheduling and CPU scheduling

J. C. Browne, Jean Lan, Forest Baskett
1972 Proceedings of the December 5-7, 1972, fall joint computer conference, part I on - AFIPS '72 (Fall, part I)  
Austin Texas December 1971 9 S SHERMAN J C BROWNE F BASKETT Trace-driven modeling and analysis of CPU scheduling in a multi-programming system To appear CACM-Also Proc of ACM Workshop on Performance Evaluation  ...  This span oLCPU utilization is in good accord with the trace-driven model results of Sherman, Baskett, and Browne. 9 It is interesting to note that for the hyper-exponentially distributed CPU service  ... 
doi:10.1145/1479992.1479995 dblp:conf/afips/BrowneLB72 fatcat:6z7vk75rgjh37hxjy66b6htsdu

The impact of job memory requirements on gang-scheduling performance

Sanjeev Setia, Mark S. Squillante, Vijay K. Naik
1999 Performance Evaluation Review  
Using trace-driven simulation, we evaluate the impact of using di erent long-term scheduling policies on the overall performance of Distributed Hierarchical Control DHC, a gang-scheduling policy that has  ...  We rst present an analysis of the memory-usage characteristics of jobs in the production workload on the Cray T3E at the San Diego Supercomputer Center.  ...  Thanks also to Morris Jette of LLNL for providing us with data on the characteristics of the ASCI workload and allowing us to report that information in this paper.  ... 
doi:10.1145/309746.309751 fatcat:x4r2x5y3q5h7jaeazzx5ho2esy

Evaluation of the memory page migration influence in the system performance

Julita Corbalan, Xavier Martorell, Jesus Labarta
2003 Proceedings of the 17th annual international conference on Supercomputing - ICS '03  
Experiments also show that the combination of performance-driven policies and the memory migration mechanism results in a system that can be automatically self-evaluated and self-configured.  ...  In particular, we have focused on the effects of the page migration mechanism on the CPU time consumed by each application, the processor allocation received, and the speedup, when applying performance-driven  ...  Results also show that performance-driven scheduling policies react better to changes in the system conditions and are more robust that those policies that do not consider the job characteristics.  ... 
doi:10.1145/782832.782833 fatcat:2iss3j5ytja4dpifvm3u47zp7a

Evaluation of the memory page migration influence in the system performance

Julita Corbalan, Xavier Martorell, Jesus Labarta
2003 Proceedings of the 17th annual international conference on Supercomputing - ICS '03  
Experiments also show that the combination of performance-driven policies and the memory migration mechanism results in a system that can be automatically self-evaluated and self-configured.  ...  In particular, we have focused on the effects of the page migration mechanism on the CPU time consumed by each application, the processor allocation received, and the speedup, when applying performance-driven  ...  Results also show that performance-driven scheduling policies react better to changes in the system conditions and are more robust that those policies that do not consider the job characteristics.  ... 
doi:10.1145/782814.782833 dblp:conf/ics/CorbalanML03 fatcat:t6bjbcuygvh33pm5y4xtkmplly

Internal Scheduling and Memory Contention

A.J. Smith
1981 IEEE Transactions on Software Engineering  
Trace driven simulations are used to verify the accuracy of the models. We find that in most cases internal scheduling has only a small effect on page fault rates and CPU utilization.  ...  We create models for memory contention in a system that uses global LRU replacement and either round robin or priority internal scheduling.  ...  Trace driven simulations using program address traces will be presented in Section IV and the results will show that our model is quite accurate.  ... 
doi:10.1109/tse.1981.230820 fatcat:crs43xhhprbzrambrwpjjs46ne

A performance model of MVS

Willy W. Chiu-We, Min Chow
1978 IBM Systems Journal  
PIO, + FIO, + SWIO PMOD can be executed in two modes, a distribution-driven mode or a trace-driven mode.  ...  The performance model, PMOD, is validated in a completely de- terministic trace-driven mode, so that we do not need to replicate model executions with several random number seeds.  ... 
doi:10.1147/sj.174.0444 fatcat:vux3skvmhve2tkxr3fh732kswe

An evaluation of parallel job scheduling for ASCI Blue-Pacific

H. Franke, J. Jann, J. E. Moreira, P. Pattnaik, M. A. Jette
1999 Proceedings of the 1999 ACM/IEEE conference on Supercomputing (CDROM) - Supercomputing '99  
In this paper we analyze the behavior of a gang-scheduling system that we are developing for the ASCI Blue-Pacific machines.  ...  Our results show that both backfilling and gang-scheduling with moderate multiprogramming levels are much more effective than simple first-come first-serve scheduling.  ...  Our performance analysis shows that, even at modest multiprogramming levels of 2 and 3, gang-scheduling can accommodate very high utilization rates of up to 90%.  ... 
doi:10.1145/331532.331577 dblp:conf/sc/FrankeJMPJ99 fatcat:wio6bwidl5a3nhyvp5xmdjiqo4

An integrated approach to parallel scheduling using gang-scheduling, backfilling, and migration

Yanyong Zhang, H. Franke, J. Moreira, A. Sivasubramaniam
2003 IEEE Transactions on Parallel and Distributed Systems  
This approach, however, can result in low system utilization and large job wait times.  ...  Effective scheduling strategies to improve response times, throughput, and utilization are an important consideration in large supercomputing environments.  ...  We started with an analysis of two commonly used strategies: backfilling and gang-scheduling.  ... 
doi:10.1109/tpds.2003.1189582 fatcat:whafvkqfljbbjnjq6rz3si227q

An Integrated Approach to Parallel Scheduling Using Gang-Scheduling, Backfilling, and Migration [chapter]

Y. Zhang, H. Franke, J. E. Moreira, A. Sivasubramaniam
2001 Lecture Notes in Computer Science  
This approach, however, can result in low system utilization and large job wait times.  ...  Effective scheduling strategies to improve response times, throughput, and utilization are an important consideration in large supercomputing environments.  ...  We started with an analysis of two commonly used strategies: backfilling and gang-scheduling.  ... 
doi:10.1007/3-540-45540-x_9 fatcat:34ldf4ptovcbjbs7aszhfyxbve

Modelling and analysis of distributed software systems

B. Kumar, Timothy A. Gonsalves
1979 Proceedings of the seventh symposium on Operating systems principles - SOSP '79  
The analysis of such models will draw heavily on the substantial work already done in the analysis of hardware models.  ...  The problem of modelling and analysing the performance of software structures for distributed computer systems is addressed.  ...  For example, if the flow through the module network and the module processing times are deterministically modelled, the analysis may be conducted using trace-driven simulation.  ... 
doi:10.1145/800215.806563 dblp:conf/sosp/KumarG79 fatcat:hz7of4mjlvebplmc5zewvjn5cm

Performance measurement tools for VM/370

P. H. Callaway
1975 IBM Systems Journal  
The monitor call instruction available on System/370 CPUs greatly facilitates the collection of event-driven trace information.  ...  high priority by the scheduler and could utilize 40 percent of the CPU of a Model 145 effectively doing no useful work.  ... 
doi:10.1147/sj.142.0134 fatcat:4z456f4awnhe7fjibzokdsnr3a

An adaptive microscheduler for a multiprogrammed computer system

Edgar M. Pass, John Gwynn
1973 Proceedings of the annual conference on - ACM'73  
In trace-driven modeling, the workload and the responses of the system to the workload are supplied as input to the model.  ...  The monitor output may be used almost without further analysis, as in the case of trace-driven modeling (Cheng, 69) .  ... 
doi:10.1145/800192.805726 dblp:conf/acm/PassG73 fatcat:fojpqfvghjcxhf5vwdrrqerewi
« Previous Showing results 1 — 15 out of 358 results