Pub Date : 1994-01-31DOI: 10.1109/MASCOT.1994.284440
D. Petriu, S. Majumdar, Jinping Lin, C. E. Hrischuk
The authors present an analytical performance model named rendezvous network with multi-threaded clients (RNMTC) for performance analysis of client-server systems. RNMTC is able to model systems with multiple clients inter-communicating with multiple servers which may represent either hardware or software system components. Each system client is described by a precedence graph, and may consist of multiple concurrent execution threads whose number can vary due to fork and join operations. The analytic method for RNMTC proposed is based on hierarchical decomposition: at the higher level the system behaviour is represented by a Markov chain (MC) model whose states correspond to all possible combinations of client execution states; at the lower level a stochastic rendezvous network (SRVN) model with simple clients corresponds to each MC slate. SRVN was previously introduced and MVA approximate analytic solutions are known. The RNMTC model has been used with a number of different test cases and the analytic results were found to be in close agreement with simulation results.<>
{"title":"Analytic performance estimation of client-server systems with multi-threaded clients","authors":"D. Petriu, S. Majumdar, Jinping Lin, C. E. Hrischuk","doi":"10.1109/MASCOT.1994.284440","DOIUrl":"https://doi.org/10.1109/MASCOT.1994.284440","url":null,"abstract":"The authors present an analytical performance model named rendezvous network with multi-threaded clients (RNMTC) for performance analysis of client-server systems. RNMTC is able to model systems with multiple clients inter-communicating with multiple servers which may represent either hardware or software system components. Each system client is described by a precedence graph, and may consist of multiple concurrent execution threads whose number can vary due to fork and join operations. The analytic method for RNMTC proposed is based on hierarchical decomposition: at the higher level the system behaviour is represented by a Markov chain (MC) model whose states correspond to all possible combinations of client execution states; at the lower level a stochastic rendezvous network (SRVN) model with simple clients corresponds to each MC slate. SRVN was previously introduced and MVA approximate analytic solutions are known. The RNMTC model has been used with a number of different test cases and the analytic results were found to be in close agreement with simulation results.<<ETX>>","PeriodicalId":288344,"journal":{"name":"Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123470573","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-01-31DOI: 10.1109/MASCOT.1994.284381
Jim Pierce, T. Mudge
The authors describe IDtrace, a binary instrumentation tool which produces execution traces for the ix86 instruction set architecture. Long execution traces can be generated quickly and easily for input to a wide variety of performance evaluation tools. Issues involved in the construction of such a tool are listed along with illustrations of the uses of different generated traces. One example observes the behavior of a branch prediction technique and another compares the cache behavior of the i486 with that of the MIPS R3000.<>
{"title":"IDtrace/spl minus/a tracing tool for i486 simulation","authors":"Jim Pierce, T. Mudge","doi":"10.1109/MASCOT.1994.284381","DOIUrl":"https://doi.org/10.1109/MASCOT.1994.284381","url":null,"abstract":"The authors describe IDtrace, a binary instrumentation tool which produces execution traces for the ix86 instruction set architecture. Long execution traces can be generated quickly and easily for input to a wide variety of performance evaluation tools. Issues involved in the construction of such a tool are listed along with illustrations of the uses of different generated traces. One example observes the behavior of a branch prediction technique and another compares the cache behavior of the i486 with that of the MIPS R3000.<<ETX>>","PeriodicalId":288344,"journal":{"name":"Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121643021","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-01-31DOI: 10.1109/MASCOT.1994.284414
Yongbing Zhang, H. Kameda, K. Shimizu
Describes three adaptive bidding load balancing algorithms, each of which balances the workload on hosts (nodes) using different amounts of system state information in a heterogeneous distributed system model. A key feature of these algorithms is that they attempt to balance the virtual delay (unfinished work) of each node. Using simulation, the algorithms are examined and compared with Eager's (1986) 'shortest algorithm', an efficient algorithm in homogeneous systems. Results show that the simple algorithms we consider in this paper, which use very little system state information, yield dramatic performance improvements over the case of no load balancing and perform much better than Eager's algorithm.<>
{"title":"Adaptive bidding load balancing algorithms in heterogeneous distributed systems","authors":"Yongbing Zhang, H. Kameda, K. Shimizu","doi":"10.1109/MASCOT.1994.284414","DOIUrl":"https://doi.org/10.1109/MASCOT.1994.284414","url":null,"abstract":"Describes three adaptive bidding load balancing algorithms, each of which balances the workload on hosts (nodes) using different amounts of system state information in a heterogeneous distributed system model. A key feature of these algorithms is that they attempt to balance the virtual delay (unfinished work) of each node. Using simulation, the algorithms are examined and compared with Eager's (1986) 'shortest algorithm', an efficient algorithm in homogeneous systems. Results show that the simple algorithms we consider in this paper, which use very little system state information, yield dramatic performance improvements over the case of no load balancing and perform much better than Eager's algorithm.<<ETX>>","PeriodicalId":288344,"journal":{"name":"Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","volume":"30 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120911870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-01-31DOI: 10.1109/MASCOT.1994.284402
N. Bergé, M. Samaan, G. Juanole, Y. Atamna
The goal of this paper is to present a methodology for modeling and analyzing the main characteristics of time critical distributed systems. This methodology is presented through the study of the fieldbus FIP and is based on the use of two complementary Petri nets based models: the labeled predicate transition nets (LPrT) which are well adapted for representing complex communication systems, but which do not consider the time characteristics; and the stochastic timed Petri nets (STPN) which have the ability to express time constraints and are then well adapted for representing time critical systems. The analysis made with the LPrT model visualises all the possible different sequences of actions. The analysis made with the STPN model shows the discriminating power of time constraints and allows to make performance evaluations. The study concerns the behavior of an application layer producer entity providing a producer user process with variable writing services and mechanisms for the temporal validity of variable values.<>
{"title":"Methodology for LAN modeling and analysis using Petri nets based models","authors":"N. Bergé, M. Samaan, G. Juanole, Y. Atamna","doi":"10.1109/MASCOT.1994.284402","DOIUrl":"https://doi.org/10.1109/MASCOT.1994.284402","url":null,"abstract":"The goal of this paper is to present a methodology for modeling and analyzing the main characteristics of time critical distributed systems. This methodology is presented through the study of the fieldbus FIP and is based on the use of two complementary Petri nets based models: the labeled predicate transition nets (LPrT) which are well adapted for representing complex communication systems, but which do not consider the time characteristics; and the stochastic timed Petri nets (STPN) which have the ability to express time constraints and are then well adapted for representing time critical systems. The analysis made with the LPrT model visualises all the possible different sequences of actions. The analysis made with the STPN model shows the discriminating power of time constraints and allows to make performance evaluations. The study concerns the behavior of an application layer producer entity providing a producer user process with variable writing services and mechanisms for the temporal validity of variable values.<<ETX>>","PeriodicalId":288344,"journal":{"name":"Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129503300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-01-31DOI: 10.1109/MASCOT.1994.284400
M. Gerke, R. Dicken, H. Hoyer
Based on a simulation system for factory workcell optimization and offline programming of kinematic systems a software interface for realtime coupling to processes is presented. Robotic components are available in the real work-cell whose controllers offer high speed interfaces for realtime path modification. Both subsystems can be connected via a communication managing computer to improve from simulation to telepresence and teleoperation. Furthermore comfortable man-machine interfaces are integrated to provide robotic control in virtual reality. Our concept depends on industrial standard components which gives easy access to practitioners.<>
{"title":"From simulation to virtual reality a robotic application","authors":"M. Gerke, R. Dicken, H. Hoyer","doi":"10.1109/MASCOT.1994.284400","DOIUrl":"https://doi.org/10.1109/MASCOT.1994.284400","url":null,"abstract":"Based on a simulation system for factory workcell optimization and offline programming of kinematic systems a software interface for realtime coupling to processes is presented. Robotic components are available in the real work-cell whose controllers offer high speed interfaces for realtime path modification. Both subsystems can be connected via a communication managing computer to improve from simulation to telepresence and teleoperation. Furthermore comfortable man-machine interfaces are integrated to provide robotic control in virtual reality. Our concept depends on industrial standard components which gives easy access to practitioners.<<ETX>>","PeriodicalId":288344,"journal":{"name":"Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","volume":"22 12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129119555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-01-31DOI: 10.1109/MASCOT.1994.284426
J. T. McHenry, S. Midkiff
This paper describes the use of VHDL for performance evaluation. VHDL simulation is coupled with support programs that supply statistical workloads and derive performance results for multicomputer networks. Models for node processing and communication using wormhole routing are described. Methods for extracting performance data from simulation results are presented. Example results are provided to illustrate the utility of the approach. For clarity, the discussion in this paper focuses on a model of a 16-node, or 4-dimensional, hypercube network. The adaptation of the models to other networks is also discussed.<>
{"title":"VHDL modeling for the performance evaluation of multicomputer networks","authors":"J. T. McHenry, S. Midkiff","doi":"10.1109/MASCOT.1994.284426","DOIUrl":"https://doi.org/10.1109/MASCOT.1994.284426","url":null,"abstract":"This paper describes the use of VHDL for performance evaluation. VHDL simulation is coupled with support programs that supply statistical workloads and derive performance results for multicomputer networks. Models for node processing and communication using wormhole routing are described. Methods for extracting performance data from simulation results are presented. Example results are provided to illustrate the utility of the approach. For clarity, the discussion in this paper focuses on a model of a 16-node, or 4-dimensional, hypercube network. The adaptation of the models to other networks is also discussed.<<ETX>>","PeriodicalId":288344,"journal":{"name":"Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122933091","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-01-31DOI: 10.1109/MASCOT.1994.284390
L. Cherkasova, A. Davis, V. Kotov, Tomas Rokicki
Large concurrent or distributed systems have proven notoriously difficult to comprehend due to their combinatoric complexity. There is a clear need for thorough formal system modeling at all design stages. Ease of use and validation capability are important criteria in method selection. It is important that the selection permit evaluation of design details with respect to performance and behavioral goals. Net models are a likely candidate to achieve the requisite needs of such a modeling environment. They provide both graphical and mathematical system modeling views which inherently enable both quantitative and qualitative analysis of the design. To demonstrate this claim, a colored Petri Net model and performance analysis of a particular scalable multiprocessor interconnect fabric is presented. The fundamental component of this fabric is an adaptive routing device, R2. The R2 fabric is based on the interconnection scheme used in the Mayfly parallel processing system.<>
{"title":"Colored Petri net methods for performance analysis of scalable high-speed interconnects","authors":"L. Cherkasova, A. Davis, V. Kotov, Tomas Rokicki","doi":"10.1109/MASCOT.1994.284390","DOIUrl":"https://doi.org/10.1109/MASCOT.1994.284390","url":null,"abstract":"Large concurrent or distributed systems have proven notoriously difficult to comprehend due to their combinatoric complexity. There is a clear need for thorough formal system modeling at all design stages. Ease of use and validation capability are important criteria in method selection. It is important that the selection permit evaluation of design details with respect to performance and behavioral goals. Net models are a likely candidate to achieve the requisite needs of such a modeling environment. They provide both graphical and mathematical system modeling views which inherently enable both quantitative and qualitative analysis of the design. To demonstrate this claim, a colored Petri Net model and performance analysis of a particular scalable multiprocessor interconnect fabric is presented. The fundamental component of this fabric is an adaptive routing device, R2. The R2 fabric is based on the interconnection scheme used in the Mayfly parallel processing system.<<ETX>>","PeriodicalId":288344,"journal":{"name":"Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124071552","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-01-31DOI: 10.1109/MASCOT.1994.284396
A. Waheed, Bernd Kronmüller, Roomi Sinha, D. Rover
Performance visualization is increasingly being considered a powerful approach for the analysis and debugging of parallel programs. Several performance analysis tools have already been developed for the analysis of program behavior. These tools have their specific advantages as well as limitations. A toolkit approach is presented that benefits from the advantages of conventional analysis tools, and also strives to overcome their limitations by giving a user the choice of a tool or analysis method that is most suitable in a particular situation. Various generic and commercially available data analysis tools contribute to the extensibility of the toolkit. One of the fundamental objectives of the toolkit is to practically implement multiple-domain analysis methods. Presently, the toolkit consists of Pablo, ParaGraph, AVS, and Matlab.<>
{"title":"A toolkit for advanced performance analysis","authors":"A. Waheed, Bernd Kronmüller, Roomi Sinha, D. Rover","doi":"10.1109/MASCOT.1994.284396","DOIUrl":"https://doi.org/10.1109/MASCOT.1994.284396","url":null,"abstract":"Performance visualization is increasingly being considered a powerful approach for the analysis and debugging of parallel programs. Several performance analysis tools have already been developed for the analysis of program behavior. These tools have their specific advantages as well as limitations. A toolkit approach is presented that benefits from the advantages of conventional analysis tools, and also strives to overcome their limitations by giving a user the choice of a tool or analysis method that is most suitable in a particular situation. Various generic and commercially available data analysis tools contribute to the extensibility of the toolkit. One of the fundamental objectives of the toolkit is to practically implement multiple-domain analysis methods. Presently, the toolkit consists of Pablo, ParaGraph, AVS, and Matlab.<<ETX>>","PeriodicalId":288344,"journal":{"name":"Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128068837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-01-31DOI: 10.1109/MASCOT.1994.284451
P. Heidelberger, D. Nicol
Consider simulating a large queueing network, i.e., one with many queues and jobs, on a parallel computer. Suppose the network is partitioned so that each processor is assigned a set of queues to simulate. The basic difficulty in parallel simulation is synchronizing the simulation clocks on each of the processors so that events appear to get executed in the proper order. If the network possesses a Markovian structure, then algorithms described by the authors (1993) can be used to simplify the clock synchronization problem. For appropriately structured queueing networks, these algorithms are quite efficient and large speedups have been obtained; in one example a speedup of 220 was measured on 256 processors of the Intel Touchstone Delta computer.<>
{"title":"Parallel simulation of Markovian queueing networks","authors":"P. Heidelberger, D. Nicol","doi":"10.1109/MASCOT.1994.284451","DOIUrl":"https://doi.org/10.1109/MASCOT.1994.284451","url":null,"abstract":"Consider simulating a large queueing network, i.e., one with many queues and jobs, on a parallel computer. Suppose the network is partitioned so that each processor is assigned a set of queues to simulate. The basic difficulty in parallel simulation is synchronizing the simulation clocks on each of the processors so that events appear to get executed in the proper order. If the network possesses a Markovian structure, then algorithms described by the authors (1993) can be used to simplify the clock synchronization problem. For appropriately structured queueing networks, these algorithms are quite efficient and large speedups have been obtained; in one example a speedup of 220 was measured on 256 processors of the Intel Touchstone Delta computer.<<ETX>>","PeriodicalId":288344,"journal":{"name":"Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132646353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-01-31DOI: 10.1109/MASCOT.1994.284433
A. Waheed, Bernd Kronmüller, D. Rover
For large-scale systems depending on numerous parameters and variables, matrices are considered to be a very useful notation. Techniques of linear algebra can be used to analyze the system and understand its behavior. To apply the matrix notation to large-scale computer systems is a novel approach to gain insight into its concurrent activities, which is not an easy task without special analysis methods. Therefore, we introduce a matrix notation for the performance data. Using a metric-based performance matrix, the transformations to the different analysis domains (e.g. spatial and temporal) are explained. This approach towards performance analysis benefits from the methods of linear algebra and provides a mathematical means of analysis as an alternative to approaches based on database systems.<>
{"title":"A matrix approach to performance data modeling, analysis and visualization","authors":"A. Waheed, Bernd Kronmüller, D. Rover","doi":"10.1109/MASCOT.1994.284433","DOIUrl":"https://doi.org/10.1109/MASCOT.1994.284433","url":null,"abstract":"For large-scale systems depending on numerous parameters and variables, matrices are considered to be a very useful notation. Techniques of linear algebra can be used to analyze the system and understand its behavior. To apply the matrix notation to large-scale computer systems is a novel approach to gain insight into its concurrent activities, which is not an easy task without special analysis methods. Therefore, we introduce a matrix notation for the performance data. Using a metric-based performance matrix, the transformations to the different analysis domains (e.g. spatial and temporal) are explained. This approach towards performance analysis benefits from the methods of linear algebra and provides a mathematical means of analysis as an alternative to approaches based on database systems.<<ETX>>","PeriodicalId":288344,"journal":{"name":"Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132997224","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：