Pub Date : 1999-04-11DOI: 10.1109/SIMSYM.1999.766465
G. Lencse
The Statistical Synchronization Method (SSM) (Gy. Pongor, 1992) is a promising alternative to the conventional synchronization methods for parallel discrete event simulation. The conditions of the statistics exchange of the Statistical Synchronization Method are studied. A statistics exchange control algorithm is proposed that is based on prediction and synchronization point deletion. The so-called penalty functions are introduced. They are used to give a mathematical criterion that can be a measure of the goodness of the different statistics exchange control algorithms. Both analytical treatment and simulation show that there is a trade-off between the accuracy of the results and the achievable speed-up. According to the simulation results, we can get near-optimal results for a relatively wide range of values of the parameters of the prediction; thus SSM-T is robust enough to tolerate some inaccuracy of the parameters of the prediction.
{"title":"Design criterion for the statistics exchange control algorithms used in the statistical synchronization method","authors":"G. Lencse","doi":"10.1109/SIMSYM.1999.766465","DOIUrl":"https://doi.org/10.1109/SIMSYM.1999.766465","url":null,"abstract":"The Statistical Synchronization Method (SSM) (Gy. Pongor, 1992) is a promising alternative to the conventional synchronization methods for parallel discrete event simulation. The conditions of the statistics exchange of the Statistical Synchronization Method are studied. A statistics exchange control algorithm is proposed that is based on prediction and synchronization point deletion. The so-called penalty functions are introduced. They are used to give a mathematical criterion that can be a measure of the goodness of the different statistics exchange control algorithms. Both analytical treatment and simulation show that there is a trade-off between the accuracy of the results and the achievable speed-up. According to the simulation results, we can get near-optimal results for a relatively wide range of values of the parameters of the prediction; thus SSM-T is robust enough to tolerate some inaccuracy of the parameters of the prediction.","PeriodicalId":104054,"journal":{"name":"Proceedings 32nd Annual Simulation Symposium","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122536343","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 : 1999-04-11DOI: 10.1109/SIMSYM.1999.766464
L. H. Tichenor, M. Edeen
A regenerative life support system simulation tool (RLSS) is being developed to provide an easy to use alternative to the CASE/A simulation environment. The RLSS package is currently in use at JSC Houston to investigate atmospheric interactions in small closed systems. The crew component of this new tool is designed to provide easy access to all of the variables and logic required to model the human component of such systems.
{"title":"The human component of an atmosphere regeneration simulation tool","authors":"L. H. Tichenor, M. Edeen","doi":"10.1109/SIMSYM.1999.766464","DOIUrl":"https://doi.org/10.1109/SIMSYM.1999.766464","url":null,"abstract":"A regenerative life support system simulation tool (RLSS) is being developed to provide an easy to use alternative to the CASE/A simulation environment. The RLSS package is currently in use at JSC Houston to investigate atmospheric interactions in small closed systems. The crew component of this new tool is designed to provide easy access to all of the variables and logic required to model the human component of such systems.","PeriodicalId":104054,"journal":{"name":"Proceedings 32nd Annual Simulation Symposium","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131716493","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 : 1999-04-11DOI: 10.1109/SIMSYM.1999.766459
M. Kourjanski, Aleks Göllü
The SHIFT programming language and simulation environment were developed at California PATH, UC Berkeley. SHIFT is a special programming language designed to model dynamical networks of hybrid automata. Immediate applications of the SHIFT simulation framework include large scale simulations for the Automated Highway Systems, air traffic control, Mobile Offshore Base concept for the US Navy. In order to increase the productivity of the simulation kernel, a parallelized version of the kernel has been developed. We show various approaches to this problem and highlight the results of parallelizing the simulation kernel.
{"title":"Parallelization of SHIFT simulation environment","authors":"M. Kourjanski, Aleks Göllü","doi":"10.1109/SIMSYM.1999.766459","DOIUrl":"https://doi.org/10.1109/SIMSYM.1999.766459","url":null,"abstract":"The SHIFT programming language and simulation environment were developed at California PATH, UC Berkeley. SHIFT is a special programming language designed to model dynamical networks of hybrid automata. Immediate applications of the SHIFT simulation framework include large scale simulations for the Automated Highway Systems, air traffic control, Mobile Offshore Base concept for the US Navy. In order to increase the productivity of the simulation kernel, a parallelized version of the kernel has been developed. We show various approaches to this problem and highlight the results of parallelizing the simulation kernel.","PeriodicalId":104054,"journal":{"name":"Proceedings 32nd Annual Simulation Symposium","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127960529","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 : 1999-04-11DOI: 10.1109/SIMSYM.1999.766449
Jong-Suk Ahn, Seunghyun Oh
Discrete event simulations need a priority queue sorting events according to their timestamp to process events in their time order. As the number of events increases, the choice of data structure for this event list can affect the simulation performance significantly. A calendar queue is a data structure popularly used in most discrete event simulators due to its O(1) time complexity regardless of the number of stored events. Calendar queues, however, perform poorly over skewed event distributions due to the static resize algorithm and the inappropriate selection of events for measuring the degree of the event distribution. To improve the calendar queue's performance over uneven event distributions, this paper proposes two new mechanisms. We call our calendar queue adopting these two mechanisms DCQ (dynamic calendar queue). Our experiment results showed that DCQ can achieve an order of magnitude speedup for uneven distributions while performing as well over even distributions as the conventional calendar queue.
{"title":"Dynamic calendar queue","authors":"Jong-Suk Ahn, Seunghyun Oh","doi":"10.1109/SIMSYM.1999.766449","DOIUrl":"https://doi.org/10.1109/SIMSYM.1999.766449","url":null,"abstract":"Discrete event simulations need a priority queue sorting events according to their timestamp to process events in their time order. As the number of events increases, the choice of data structure for this event list can affect the simulation performance significantly. A calendar queue is a data structure popularly used in most discrete event simulators due to its O(1) time complexity regardless of the number of stored events. Calendar queues, however, perform poorly over skewed event distributions due to the static resize algorithm and the inappropriate selection of events for measuring the degree of the event distribution. To improve the calendar queue's performance over uneven event distributions, this paper proposes two new mechanisms. We call our calendar queue adopting these two mechanisms DCQ (dynamic calendar queue). Our experiment results showed that DCQ can achieve an order of magnitude speedup for uneven distributions while performing as well over even distributions as the conventional calendar queue.","PeriodicalId":104054,"journal":{"name":"Proceedings 32nd Annual Simulation Symposium","volume":"39 9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128335358","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 : 1999-04-11DOI: 10.1109/SIMSYM.1999.766462
Lamine Ngom, C. Baron, J. Geffroy
Identification methods (formal or simulation based), are used for logical design, test or sequential learning. Roughly, we can say that they consist of deriving an automaton model of a given sequential system from a functional description of its behavior. We present a novel identification approach based on genetic simulation. The first section offers a synthetic unified classification of the different known identification methods according to three criteria that have been extracted from their analysis. Then, the potentiality and interest of genetic simulation for identification is analyzed and a new genetic approach for functional identification is presented. Lastly we describe a computational experiment we made to validate our idea and the results we obtained. New perspectives are wide open now, particularly concerning the design, simulation and behavioral prediction of incremental and adaptive systems.
{"title":"Genetic simulation for finite state machine identification","authors":"Lamine Ngom, C. Baron, J. Geffroy","doi":"10.1109/SIMSYM.1999.766462","DOIUrl":"https://doi.org/10.1109/SIMSYM.1999.766462","url":null,"abstract":"Identification methods (formal or simulation based), are used for logical design, test or sequential learning. Roughly, we can say that they consist of deriving an automaton model of a given sequential system from a functional description of its behavior. We present a novel identification approach based on genetic simulation. The first section offers a synthetic unified classification of the different known identification methods according to three criteria that have been extracted from their analysis. Then, the potentiality and interest of genetic simulation for identification is analyzed and a new genetic approach for functional identification is presented. Lastly we describe a computational experiment we made to validate our idea and the results we obtained. New perspectives are wide open now, particularly concerning the design, simulation and behavioral prediction of incremental and adaptive systems.","PeriodicalId":104054,"journal":{"name":"Proceedings 32nd Annual Simulation Symposium","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132266503","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 : 1999-04-11DOI: 10.1109/SIMSYM.1999.766461
O. Castillo, P. Melin
We describe a novel method for mathematical modelling and simulation (MMS) of nonlinear robotic dynamic systems using fuzzy logic techniques, genetic algorithms and fractal theory. The new fuzzy-fractal-genetic method combines soft computing techniques with mathematical methods for the domain of modelling and simulation of nonlinear robotic systems. This domain is quite complex because it is a well known fact that even simple nonlinear dynamical systems can exhibit "chaotic" behavior. The new method for MMS of nonlinear robotic dynamic systems has been implemented as a computer program to show that our new fuzzy-fractal-genetic approach is a good alternative for modelling this kind of system.
{"title":"Intelligent mathematical modelling and simulation of robotic dynamic systems using a new fuzzy-fractal-genetic approach","authors":"O. Castillo, P. Melin","doi":"10.1109/SIMSYM.1999.766461","DOIUrl":"https://doi.org/10.1109/SIMSYM.1999.766461","url":null,"abstract":"We describe a novel method for mathematical modelling and simulation (MMS) of nonlinear robotic dynamic systems using fuzzy logic techniques, genetic algorithms and fractal theory. The new fuzzy-fractal-genetic method combines soft computing techniques with mathematical methods for the domain of modelling and simulation of nonlinear robotic systems. This domain is quite complex because it is a well known fact that even simple nonlinear dynamical systems can exhibit \"chaotic\" behavior. The new method for MMS of nonlinear robotic dynamic systems has been implemented as a computer program to show that our new fuzzy-fractal-genetic approach is a good alternative for modelling this kind of system.","PeriodicalId":104054,"journal":{"name":"Proceedings 32nd Annual Simulation Symposium","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132453030","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 : 1999-04-11DOI: 10.1109/SIMSYM.1999.766466
E. P. O'Grady
An algorithm suited for hardware implementation assists in evaluating two-variable floating-point map functions. A map function F(X,Z) is an arbitrary function found in jet engine simulations and other scientific applications. The hardware oriented algorithm involves two stages of processing: preprocessing and run time processing. During preprocessing, breakpoints are analyzed to generate data tables and constants used at run time. At run time, the breakpoint region containing an arbitrary input argument (X/spl nu/,Z/spl nu/), is identified in a small number of steps. The algorithm replaces the general search needed in software routines to identify the breakpoint region. An example describes a straightforward implementation of a map function generator that uses bilinear interpolation.
{"title":"Hardware support for floating point map function generation","authors":"E. P. O'Grady","doi":"10.1109/SIMSYM.1999.766466","DOIUrl":"https://doi.org/10.1109/SIMSYM.1999.766466","url":null,"abstract":"An algorithm suited for hardware implementation assists in evaluating two-variable floating-point map functions. A map function F(X,Z) is an arbitrary function found in jet engine simulations and other scientific applications. The hardware oriented algorithm involves two stages of processing: preprocessing and run time processing. During preprocessing, breakpoints are analyzed to generate data tables and constants used at run time. At run time, the breakpoint region containing an arbitrary input argument (X/spl nu/,Z/spl nu/), is identified in a small number of steps. The algorithm replaces the general search needed in software routines to identify the breakpoint region. An example describes a straightforward implementation of a map function generator that uses bilinear interpolation.","PeriodicalId":104054,"journal":{"name":"Proceedings 32nd Annual Simulation Symposium","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130845725","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 : 1999-04-11DOI: 10.1109/SIMSYM.1999.766458
Y. M. Teo, S. Tay
Developing a parallel discrete event simulation from scratch requires an in-depth knowledge of the mapping process from the physical model to the simulation model, and a substantial effort in coping with numerous parallelism issues in the underlying synchronization protocols adopted. The lack of software tools and environments to reduce the development effort significantly is a major hindrance in adopting parallel simulation technology. The paper presents an overview of the SPaDES (Structured Parallel Discrete-Event Simulation) scalable parallel simulation framework. We focus on the performance analysis of SPaDES/C/sub ++/, an implementation of SPaDES on a distributed memory Fujitsu AP3000 parallel computer. SPaDES/C/sub ++/ hides the underlying complex parallel simulation synchronization and parallel programming details from the simulationist. We study various ways of improving SPaDES execution performance including periodic checkpointing of simulation states, aggregation of messages for logical processes that reside on the same physical processors, and increasing the computational granularity of run time processes to reduce the costs of synchronization and communication. Our empirical results show that the SPaDES framework can deliver good speedup for applications with large problem size and is scalable.
{"title":"Performance evaluation of a parallel simulation environment","authors":"Y. M. Teo, S. Tay","doi":"10.1109/SIMSYM.1999.766458","DOIUrl":"https://doi.org/10.1109/SIMSYM.1999.766458","url":null,"abstract":"Developing a parallel discrete event simulation from scratch requires an in-depth knowledge of the mapping process from the physical model to the simulation model, and a substantial effort in coping with numerous parallelism issues in the underlying synchronization protocols adopted. The lack of software tools and environments to reduce the development effort significantly is a major hindrance in adopting parallel simulation technology. The paper presents an overview of the SPaDES (Structured Parallel Discrete-Event Simulation) scalable parallel simulation framework. We focus on the performance analysis of SPaDES/C/sub ++/, an implementation of SPaDES on a distributed memory Fujitsu AP3000 parallel computer. SPaDES/C/sub ++/ hides the underlying complex parallel simulation synchronization and parallel programming details from the simulationist. We study various ways of improving SPaDES execution performance including periodic checkpointing of simulation states, aggregation of messages for logical processes that reside on the same physical processors, and increasing the computational granularity of run time processes to reduce the costs of synchronization and communication. Our empirical results show that the SPaDES framework can deliver good speedup for applications with large problem size and is scalable.","PeriodicalId":104054,"journal":{"name":"Proceedings 32nd Annual Simulation Symposium","volume":"106 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128603544","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 : 1999-04-11DOI: 10.1109/SIMSYM.1999.766455
D. Parson, Paul Beatty, C. Glossner, Bryan Schlieder
The paper examines the layered software modules of a heterogeneous multiprocessor simulator and debugger, and the design patterns that span these modules. Lucent's LUxWORKS simulator and debugger, works with multiple processor architectures. Its modeling infrastructure, processor models, processor monitor/control, hardware control, vendor simulator interface and Tcl/Tk extension layers are spanned by the following design patterns: (1) build and extend abstract virtual processors; (2) build reflective entities; and (3) build a covariant extensible system. Together these modules and patterns define a processor execution architecture that encourages reuse and dynamic extensibility.
{"title":"A framework for simulating heterogeneous virtual processors","authors":"D. Parson, Paul Beatty, C. Glossner, Bryan Schlieder","doi":"10.1109/SIMSYM.1999.766455","DOIUrl":"https://doi.org/10.1109/SIMSYM.1999.766455","url":null,"abstract":"The paper examines the layered software modules of a heterogeneous multiprocessor simulator and debugger, and the design patterns that span these modules. Lucent's LUxWORKS simulator and debugger, works with multiple processor architectures. Its modeling infrastructure, processor models, processor monitor/control, hardware control, vendor simulator interface and Tcl/Tk extension layers are spanned by the following design patterns: (1) build and extend abstract virtual processors; (2) build reflective entities; and (3) build a covariant extensible system. Together these modules and patterns define a processor execution architecture that encourages reuse and dynamic extensibility.","PeriodicalId":104054,"journal":{"name":"Proceedings 32nd Annual Simulation Symposium","volume":"149 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122344723","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 : 1999-04-11DOI: 10.1109/SIMSYM.1999.766450
H. Karatza
In distributed systems job scheduling is a difficult problem. In this work we study a special type of scheduling called gang scheduling under which jobs consist of a number of interacting tasks which are scheduled to run simultaneously on distinct processors. The performance of various gang scheduling schemes is studied and compared for a variety of workloads. The main objective of the processor schedulers is to achieve high overall system throughput while at the same time providing some guarantee for the performance of individual jobs in the system.
{"title":"A simulation-based performance analysis of gang scheduling in a distributed system","authors":"H. Karatza","doi":"10.1109/SIMSYM.1999.766450","DOIUrl":"https://doi.org/10.1109/SIMSYM.1999.766450","url":null,"abstract":"In distributed systems job scheduling is a difficult problem. In this work we study a special type of scheduling called gang scheduling under which jobs consist of a number of interacting tasks which are scheduled to run simultaneously on distinct processors. The performance of various gang scheduling schemes is studied and compared for a variety of workloads. The main objective of the processor schedulers is to achieve high overall system throughput while at the same time providing some guarantee for the performance of individual jobs in the system.","PeriodicalId":104054,"journal":{"name":"Proceedings 32nd Annual Simulation Symposium","volume":"358 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125643818","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：