Pub Date : 2002-04-14DOI: 10.1109/SIMSYM.2002.1000164
M. Hybinette, R. Fujimoto
In previous work we presented an algorithm for cloning parallel simulations that enables multiple simulated execution paths to be explored simultaneously. The method is targeted for parallel discrete event simulators that provide the simulation application developer a logical process (LP) execution model. The cloning algorithm gains efficiency by cloning logical processes only as necessary. In this work we examine the scalability of cloning in detail. Specifically, we examine how the number of clones impacts the performance of cloning as we vary the "size" of the simulation problem.
{"title":"Scalability of parallel simulation cloning","authors":"M. Hybinette, R. Fujimoto","doi":"10.1109/SIMSYM.2002.1000164","DOIUrl":"https://doi.org/10.1109/SIMSYM.2002.1000164","url":null,"abstract":"In previous work we presented an algorithm for cloning parallel simulations that enables multiple simulated execution paths to be explored simultaneously. The method is targeted for parallel discrete event simulators that provide the simulation application developer a logical process (LP) execution model. The cloning algorithm gains efficiency by cloning logical processes only as necessary. In this work we examine the scalability of cloning in detail. Specifically, we examine how the number of clones impacts the performance of cloning as we vary the \"size\" of the simulation problem.","PeriodicalId":198576,"journal":{"name":"Proceedings 35th Annual Simulation Symposium. SS 2002","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127581920","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 : 2002-04-14DOI: 10.1109/SIMSYM.2002.1000079
Irina Gerasimov, R. Simon
One of the major challenges faced by designers of ad hoc network systems is the deployment of end-to-end quality-of-service support mechanisms. This paper describes QoS-AODV, an integrated route discovery and bandwidth reservation protocol. QoS-AODV is designed to operate within a TDMA network. Unlike other path finding and route discovery protocols that ignore the impact of the data link layer, QoS-AODV incorporates slot scheduling information to ensure that end-to-end bandwidth is actually reserved. Further, QoS-AODV is an enhanced version of the Ad hoc On-demand Distance Vector routing protocol, and is therefore compatible with proposed route discovery and maintenance techniques. In order to test the effectiveness of our protocol we implemented a version of QoS-AODV, along with two similar protocols, in the ns-2 simulator. Our experiments showed the QoS-AODV significantly improved the probability of being able to find an end-to-end QoS path.
{"title":"A bandwidth-reservation mechanism for on-demand ad hoc path finding","authors":"Irina Gerasimov, R. Simon","doi":"10.1109/SIMSYM.2002.1000079","DOIUrl":"https://doi.org/10.1109/SIMSYM.2002.1000079","url":null,"abstract":"One of the major challenges faced by designers of ad hoc network systems is the deployment of end-to-end quality-of-service support mechanisms. This paper describes QoS-AODV, an integrated route discovery and bandwidth reservation protocol. QoS-AODV is designed to operate within a TDMA network. Unlike other path finding and route discovery protocols that ignore the impact of the data link layer, QoS-AODV incorporates slot scheduling information to ensure that end-to-end bandwidth is actually reserved. Further, QoS-AODV is an enhanced version of the Ad hoc On-demand Distance Vector routing protocol, and is therefore compatible with proposed route discovery and maintenance techniques. In order to test the effectiveness of our protocol we implemented a version of QoS-AODV, along with two similar protocols, in the ns-2 simulator. Our experiments showed the QoS-AODV significantly improved the probability of being able to find an end-to-end QoS path.","PeriodicalId":198576,"journal":{"name":"Proceedings 35th Annual Simulation Symposium. SS 2002","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130023302","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 : 2002-04-14DOI: 10.1109/SIMSYM.2002.1000088
R. Beraldi, L. Nigro, A. Orlando, F. Pupo
This paper introduces TUTW - Temporal Uncertainty Time Warp - a control engine designed for an exploitation of temporal uncertainty (TU) in general optimistic simulations, and concentrates on an agent-based implementation of TUTW which enables distributed simulations over the Internet. A novel concept in TUTW is an event model where time intervals are attached to events instead of classical punctual or "precise" timestamps. All of this reflects typical system specifications where the occurrence time of events cannot be known with complete certainty. TUTW takes advantage of TU by resolving events so that the number of rollbacks is reduced. The simulator performance can thus be improved without necessarily compromising the accuracy of the results. The current agent-based implementation is based on Java and the ActorFoundry middleware. The implementation is totally portable and makes it possible to configure a distributed simulation virtual machine which includes heterogeneous computing platforms. The paper includes an example to illustrate the application of TUTW.
本文介绍了TUTW - Temporal Uncertainty Time Warp -一个为利用一般乐观模拟中的时间不确定性(TU)而设计的控制引擎，并重点介绍了TUTW的基于代理的实现，该实现使Internet上的分布式模拟成为可能。TUTW中的一个新概念是事件模型，其中时间间隔附加到事件上，而不是传统的准时或“精确”时间戳。所有这些都反映了典型的系统规范，其中事件的发生时间不能完全确定。TUTW通过解析事件来利用TU，从而减少了回滚的次数。因此，可以在不损害结果准确性的情况下改进模拟器的性能。当前基于代理的实现是基于Java和ActorFoundry中间件的。该实现是完全可移植的，并且使得配置包含异构计算平台的分布式仿真虚拟机成为可能。本文通过实例说明了TUTW的应用。
{"title":"Temporal uncertainty time warp: an agent-based implementation","authors":"R. Beraldi, L. Nigro, A. Orlando, F. Pupo","doi":"10.1109/SIMSYM.2002.1000088","DOIUrl":"https://doi.org/10.1109/SIMSYM.2002.1000088","url":null,"abstract":"This paper introduces TUTW - Temporal Uncertainty Time Warp - a control engine designed for an exploitation of temporal uncertainty (TU) in general optimistic simulations, and concentrates on an agent-based implementation of TUTW which enables distributed simulations over the Internet. A novel concept in TUTW is an event model where time intervals are attached to events instead of classical punctual or \"precise\" timestamps. All of this reflects typical system specifications where the occurrence time of events cannot be known with complete certainty. TUTW takes advantage of TU by resolving events so that the number of rollbacks is reduced. The simulator performance can thus be improved without necessarily compromising the accuracy of the results. The current agent-based implementation is based on Java and the ActorFoundry middleware. The implementation is totally portable and makes it possible to configure a distributed simulation virtual machine which includes heterogeneous computing platforms. The paper includes an example to illustrate the application of TUTW.","PeriodicalId":198576,"journal":{"name":"Proceedings 35th Annual Simulation Symposium. SS 2002","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116307707","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 : 2002-04-14DOI: 10.1109/SIMSYM.2002.1000167
E. Danahy, S. A. Morrison
Imagine using a virtual learning environment to remove the programmer from the process of developing code for mechanical movement. Efficient artificial intelligence combined with a high fidelity simulation would allow the computer to discover valid, optimal actions for a robot in faster than real-time, thus eliminating the need for human guess-and-test. This paper presents the challenges of developing such a system, and describes a robotic machine and associated simulation that gives testimony to this possibility.
{"title":"Faster than real-time machine learning within high fidelity simulations","authors":"E. Danahy, S. A. Morrison","doi":"10.1109/SIMSYM.2002.1000167","DOIUrl":"https://doi.org/10.1109/SIMSYM.2002.1000167","url":null,"abstract":"Imagine using a virtual learning environment to remove the programmer from the process of developing code for mechanical movement. Efficient artificial intelligence combined with a high fidelity simulation would allow the computer to discover valid, optimal actions for a robot in faster than real-time, thus eliminating the need for human guess-and-test. This paper presents the challenges of developing such a system, and describes a robotic machine and associated simulation that gives testimony to this possibility.","PeriodicalId":198576,"journal":{"name":"Proceedings 35th Annual Simulation Symposium. SS 2002","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129476939","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 : 2002-04-14DOI: 10.1109/SIMSYM.2002.1000081
L. Kant
The use of wireless communications networks has gained phenomenal momentum both in the business and individual-user sectors. As a consequence of the increasing reliance of businesses on such networks, service disruption due to network facility failures can cause severe economic impact on the associated business. Hence service survivability-i.e., providing continued service in the midst of network failures becomes an important feature for network providers to incorporate within their network. In this work, we propose appropriate self-healing mechanisms that can be used by network providers/engineers to provide automatic service restoration (hence service survivability) in the event of network facility failures in the wireless networks of today. In particular, our contribution is as follows. We propose appropriate (cost-performance- and complexity-sensitive) self-healing mechanisms that can be used both in the currently deployed GSM-based and the rapidly emerging GPRS-based wireless networks to provide service survivability. We then perform modeling and simulation of portions of the proposed self-healing mechanisms and analyze the performance of the proposed self-healing mechanism in the core network portion of a GSM/GPRS-based system. Our simulations not only help to demonstrate the importance of simulation-based studies towards understanding, modeling and analyzing the performance of the proposed self healing mechanisms but also help provide useful insights into the potential of the proposed self-healing mechanisms.
{"title":"Design and performance modeling & simulation of self-healing mechanisms for wireless communication networks","authors":"L. Kant","doi":"10.1109/SIMSYM.2002.1000081","DOIUrl":"https://doi.org/10.1109/SIMSYM.2002.1000081","url":null,"abstract":"The use of wireless communications networks has gained phenomenal momentum both in the business and individual-user sectors. As a consequence of the increasing reliance of businesses on such networks, service disruption due to network facility failures can cause severe economic impact on the associated business. Hence service survivability-i.e., providing continued service in the midst of network failures becomes an important feature for network providers to incorporate within their network. In this work, we propose appropriate self-healing mechanisms that can be used by network providers/engineers to provide automatic service restoration (hence service survivability) in the event of network facility failures in the wireless networks of today. In particular, our contribution is as follows. We propose appropriate (cost-performance- and complexity-sensitive) self-healing mechanisms that can be used both in the currently deployed GSM-based and the rapidly emerging GPRS-based wireless networks to provide service survivability. We then perform modeling and simulation of portions of the proposed self-healing mechanisms and analyze the performance of the proposed self-healing mechanism in the core network portion of a GSM/GPRS-based system. Our simulations not only help to demonstrate the importance of simulation-based studies towards understanding, modeling and analyzing the performance of the proposed self healing mechanisms but also help provide useful insights into the potential of the proposed self-healing mechanisms.","PeriodicalId":198576,"journal":{"name":"Proceedings 35th Annual Simulation Symposium. SS 2002","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124031290","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 : 2002-04-14DOI: 10.1109/SIMSYM.2002.1000174
P. Fritzson, Peter Bunus
Modelica is a general equation-based object-oriented language for continuous and discrete-event modeling of physical systems for the purpose of efficient simulation. The language unifies and generalizes previous object-oriented modeling languages. It is bringing about a revolution in this area, based on its ease of use, visual design of models with combination of Lego-like predefined model building blocks, its ability to define model libraries with re-usable components and its support for modeling and simulation of complex applications involving parts from several application domains. In this paper we present the Modelica language with emphasis on its language features and one of the associated simulation environments. Simulation models can be developed in an integrated problem-solving environment by using a graphical editor for connection diagrams. Connections are established just by drawing lines between objects picked from a class library. The principles of object oriented physical systems modeling and the multi-domain capabilities of the language are presented in the paper by several examples.
{"title":"Modelica - a general object-oriented language for continuous and discrete-event system modeling and simulation","authors":"P. Fritzson, Peter Bunus","doi":"10.1109/SIMSYM.2002.1000174","DOIUrl":"https://doi.org/10.1109/SIMSYM.2002.1000174","url":null,"abstract":"Modelica is a general equation-based object-oriented language for continuous and discrete-event modeling of physical systems for the purpose of efficient simulation. The language unifies and generalizes previous object-oriented modeling languages. It is bringing about a revolution in this area, based on its ease of use, visual design of models with combination of Lego-like predefined model building blocks, its ability to define model libraries with re-usable components and its support for modeling and simulation of complex applications involving parts from several application domains. In this paper we present the Modelica language with emphasis on its language features and one of the associated simulation environments. Simulation models can be developed in an integrated problem-solving environment by using a graphical editor for connection diagrams. Connections are established just by drawing lines between objects picked from a class library. The principles of object oriented physical systems modeling and the multi-domain capabilities of the language are presented in the paper by several examples.","PeriodicalId":198576,"journal":{"name":"Proceedings 35th Annual Simulation Symposium. SS 2002","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132058879","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 : 2002-04-14DOI: 10.1109/SIMSYM.2002.1000156
R. Chamberlain, Ch'ng Shi Baw, M. Franklin, C. Hackmann, P. Krishnamurthy, Abhijit Mahajan, M. Wrighton
This paper describes the design and use of the interconnection network simulator (ICNS) framework. ICNS is a modular, object-oriented simulation system that has been developed to investigate performance issues in multiprocessor interconnection networks that exploit photonic technology in their design. We describe the ICNS infrastructure, present two distinct photonic interconnection networks that have been modeled using ICNS, and give performance results for each of these networks.
{"title":"Evaluating the performance of photonic interconnection networks","authors":"R. Chamberlain, Ch'ng Shi Baw, M. Franklin, C. Hackmann, P. Krishnamurthy, Abhijit Mahajan, M. Wrighton","doi":"10.1109/SIMSYM.2002.1000156","DOIUrl":"https://doi.org/10.1109/SIMSYM.2002.1000156","url":null,"abstract":"This paper describes the design and use of the interconnection network simulator (ICNS) framework. ICNS is a modular, object-oriented simulation system that has been developed to investigate performance issues in multiprocessor interconnection networks that exploit photonic technology in their design. We describe the ICNS infrastructure, present two distinct photonic interconnection networks that have been modeled using ICNS, and give performance results for each of these networks.","PeriodicalId":198576,"journal":{"name":"Proceedings 35th Annual Simulation Symposium. SS 2002","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116852671","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 : 2002-04-14DOI: 10.1109/SIMSYM.2002.1000169
Huifen Chen
We consider the problem of computing multiple headways for a single bus line to maximize the expected daily profit. The stochastic bus-line model assumes that (1) the passenger arrivals follow a Poisson process with possible reneging; (2) the number of alighting passengers at each stop follows a binomial distribution; and (3) the bus travel time follows a Weibull distribution. The objective function - the expected daily profit, defined as the ticket revenue minus the operating and customer waiting costs - is discontinuous at changes in the bus frequency. For this stochastic optimization problem, we propose a retrospective optimization algorithm that can handle both homogeneous and nonhomogeneous Poisson arrivals. Simulation results are discussed.
{"title":"Stochastic optimization in computing multiple headways for a single bus line","authors":"Huifen Chen","doi":"10.1109/SIMSYM.2002.1000169","DOIUrl":"https://doi.org/10.1109/SIMSYM.2002.1000169","url":null,"abstract":"We consider the problem of computing multiple headways for a single bus line to maximize the expected daily profit. The stochastic bus-line model assumes that (1) the passenger arrivals follow a Poisson process with possible reneging; (2) the number of alighting passengers at each stop follows a binomial distribution; and (3) the bus travel time follows a Weibull distribution. The objective function - the expected daily profit, defined as the ticket revenue minus the operating and customer waiting costs - is discontinuous at changes in the bus frequency. For this stochastic optimization problem, we propose a retrospective optimization algorithm that can handle both homogeneous and nonhomogeneous Poisson arrivals. Simulation results are discussed.","PeriodicalId":198576,"journal":{"name":"Proceedings 35th Annual Simulation Symposium. SS 2002","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124370996","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 : 2002-04-14DOI: 10.1109/SIMSYM.2002.1000078
A. Boukerche, V. Sheetal, Myongsu Choe
Ad hoc networks are useful for providing communication support where no fixed infrastructure exists or the deployment of a fixed infrastructure is not economically profitable, and movement of communicating parties is allowed. Therefore, such networks are designed to operate in widely varying environments, from military networks to low-power sensor networks and other embedded systems. Frequent topology changes caused by node mobility make routing in ad hoc wireless networks a challenging problem. In this paper, we propose an optimization technique, which we refer to as GDSR, a reactive protocol that makes use of DSR scheme and the Global positioning system (GPS). As opposed to the DSR protocol, our GDSR scheme consists of propagating the route request messages only to the nodes that are further away from the query source. We discuss the algorithm, its implementation and present an extensive simulation and experimental results to study its performance. We also present a comparative study of GDSR protocol with the existing DSR protocol. Our results clearly indicate that the GDSR protocol outperforms the DSR protocol by significantly decreasing the number of route query packets thereby increasing the efficiency of the network load. Further more, we show that a careful GPS screening angle is an important factor in the success of GDSR ad hoc routing protocol.
{"title":"A route discovery optimization scheme using GPS system","authors":"A. Boukerche, V. Sheetal, Myongsu Choe","doi":"10.1109/SIMSYM.2002.1000078","DOIUrl":"https://doi.org/10.1109/SIMSYM.2002.1000078","url":null,"abstract":"Ad hoc networks are useful for providing communication support where no fixed infrastructure exists or the deployment of a fixed infrastructure is not economically profitable, and movement of communicating parties is allowed. Therefore, such networks are designed to operate in widely varying environments, from military networks to low-power sensor networks and other embedded systems. Frequent topology changes caused by node mobility make routing in ad hoc wireless networks a challenging problem. In this paper, we propose an optimization technique, which we refer to as GDSR, a reactive protocol that makes use of DSR scheme and the Global positioning system (GPS). As opposed to the DSR protocol, our GDSR scheme consists of propagating the route request messages only to the nodes that are further away from the query source. We discuss the algorithm, its implementation and present an extensive simulation and experimental results to study its performance. We also present a comparative study of GDSR protocol with the existing DSR protocol. Our results clearly indicate that the GDSR protocol outperforms the DSR protocol by significantly decreasing the number of route query packets thereby increasing the efficiency of the network load. Further more, we show that a careful GPS screening angle is an important factor in the success of GDSR ad hoc routing protocol.","PeriodicalId":198576,"journal":{"name":"Proceedings 35th Annual Simulation Symposium. SS 2002","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124826967","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 : 2002-04-14DOI: 10.1109/SIMSYM.2002.1000160
Y. M. Teo, Y. Ng
We describe the design, implementation and performance optimizations of SPaDES/Java, a process-oriented discrete-event simulation library in Java that supports sequential and parallel simulation. Parallel event synchronization is facilitated through a hybrid carrier-null, demand-driven flushing conservative null message mechanism. Inter-processor message communication is coordinated by a shared persistent memory implemented using Java Jini/JavaSpaces. We present the stepwise performance optimizations we have carried out, focusing mainly on reducing the cost of multithreading, null message synchronization overhead, and the cost of inter-processor communication. Two benchmark programs consisting of an open linear pipeline system and PHOLD representing a closed system are used. For PHOLD(16/spl times/6, 16), our optimization reduces the null message ratio from 0.94 to 0.29 on eight processors. Based on our time and space instrumentation, we observed that the memory cost of null message synchronization accounts for less than 10% of the total memory required by the PHOLD simulation.
{"title":"SPaDES/Java: object-oriented parallel discrete-event simulation","authors":"Y. M. Teo, Y. Ng","doi":"10.1109/SIMSYM.2002.1000160","DOIUrl":"https://doi.org/10.1109/SIMSYM.2002.1000160","url":null,"abstract":"We describe the design, implementation and performance optimizations of SPaDES/Java, a process-oriented discrete-event simulation library in Java that supports sequential and parallel simulation. Parallel event synchronization is facilitated through a hybrid carrier-null, demand-driven flushing conservative null message mechanism. Inter-processor message communication is coordinated by a shared persistent memory implemented using Java Jini/JavaSpaces. We present the stepwise performance optimizations we have carried out, focusing mainly on reducing the cost of multithreading, null message synchronization overhead, and the cost of inter-processor communication. Two benchmark programs consisting of an open linear pipeline system and PHOLD representing a closed system are used. For PHOLD(16/spl times/6, 16), our optimization reduces the null message ratio from 0.94 to 0.29 on eight processors. Based on our time and space instrumentation, we observed that the memory cost of null message synchronization accounts for less than 10% of the total memory required by the PHOLD simulation.","PeriodicalId":198576,"journal":{"name":"Proceedings 35th Annual Simulation Symposium. SS 2002","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129880313","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：