The field of parallel discrete event simulation is entering a period of self-assessment. Fifteen years of investigation has witnessed great strides in techniques for efficiently executing discrete event simulations on...
详细信息
ISBN:
(纸本)1565550277
The field of parallel discrete event simulation is entering a period of self-assessment. Fifteen years of investigation has witnessed great strides in techniques for efficiently executing discrete event simulations on parallel and distributed machines. Still, the discrete event simulation community at large has failed to recognize many of these results. The central question is, why has this occurred? One possible reason is an apparent disagreement in both the focus and objectives of the parallel discrete event simulation research community (primarily computer scientists) and the discrete event simulation community (a widely diverse group including, operations researchers, management scientists, mathematicians, and statisticians, as well as computer scientists). An examination of parallel discrete event simulation from a modeling methodological perspective illustrates some of these differences and reveals potentials for their resolution.
Multi-resolution representation of simulated entities is considered essential for a growing portion of distributedsimulations. Heretofore, modelers have represented entities at just one level of resolution, or have r...
详细信息
Multi-resolution representation of simulated entities is considered essential for a growing portion of distributedsimulations. Heretofore, modelers have represented entities at just one level of resolution, or have represented concurrent representations in an inconsistent manner. We address the question of the cost of maintaining multiple, concurrent representations. We present a brief overview of our concept of a Multiple Resolution Entity (MRE) and Attribute Dependency Graph (ADG) both originally described elsewhere, and then compare simulation and consistency costs of some approaches, including our own MRE/ADG-based approach, to multi-resolution modeling. The cost analysis presented here is the first known analysis of its type, and will provide a basis for simulation designers to determine the best, and most cost-effective approach to supporting simulation of entities at different levels of resolution concurrently.
The proceedings contain 30 papers. The topics discussed include: improving lookahead in parallel multiprocessor simulation using dynamic execution path prediction;the distributed open network emulator: using relativis...
详细信息
ISBN:
(纸本)0769525873
The proceedings contain 30 papers. The topics discussed include: improving lookahead in parallel multiprocessor simulation using dynamic execution path prediction;the distributed open network emulator: using relativistic time for distributed scalable simulation;performance analysis of shared data access algorithms for distributedsimulation of multi-agent systems;predicting performance of resolution changes in parallelsimulations;what comes after the semantic web? PADS implications for the dynamic web;discrete-event execution alternatives on general purpose graphical processing units;developing an hierarchical simulator for Beta-binders;greedy algorithms for client assignment in large-scale distributed virtual environments;and interest operators: facilitating attribute interest criteria for formula-based interest management in distributed virtual environments.
Discrete-event simulation is an important tool used for the performance evaluation of parallel systems. The space of tradeoffs is large however, when attempting to balance model fidelity and simulation execution time....
详细信息
Discrete-event simulation is an important tool used for the performance evaluation of parallel systems. The space of tradeoffs is large however, when attempting to balance model fidelity and simulation execution time. This paper describes a simulator - TAPS (Threaded Application parallel System Simulator) - that, in the context of threaded parallel computations, provides a spectrum of possibilities in this tradeoff space. TAPS is specifically designed to be parallelized;we discuss some crucial considerations regarding its parallelization.
This paper presents two new versions of the Critical Channel Traversing (CCT) algorithm. CCT is a conservative parallel discrete event simulation algorithm that has been shown to achieve very high performance when use...
详细信息
ISBN:
(纸本)0769516084
This paper presents two new versions of the Critical Channel Traversing (CCT) algorithm. CCT is a conservative parallel discrete event simulation algorithm that has been shown to achieve very high performance when used in a wide area computer network simulator The first of the new algorithms called simple sender side CCT is similar to the original, but busy waiting is eliminated. Results presented show that simple sender side CCT avoids performance problems that can be caused by busy waiting. The second new algorithm called receive side CCT employs a different strategy for updating channel clocks and determining which objects should be scheduled on critical channels. Performance results show that this version provides better scaling with respect to the connectivity of the model, at the expense of some added complexity.
Federated simulation interfaces such us the High Level Architecture (HLA) were designed for interoperability, and as such are not traditionally associated with high-performance computing. In this paper, we present res...
详细信息
ISBN:
(纸本)0769519709
Federated simulation interfaces such us the High Level Architecture (HLA) were designed for interoperability, and as such are not traditionally associated with high-performance computing. In this paper, we present results of a case study examining the use of federated simulations using runtime infrastructure (RTI) software to realize large-scale parallel network simulators. We examine the performance of two different federated network simulators, and describe RTI performance optimizations that were used to achieve efficient execution. We show that RTI-based parallel simulators can scale extremely well and achieve very high *** experiments yielded more than 80-fold scaled speedup in simulating large TCP/IP networks, demonstrating performance of up to 6 million simulated packet transmissions per second on a Linux cluster. Networks containing up to two million network nodes (routers and end systems) were simulated.
The proceedings contain 81 papers. The topics discussed include: a new approach to concurrent ring: 1 bit latency;image processing on parallel machines: a protocol for managing global objects;representation and measur...
ISBN:
(纸本)0818653701
The proceedings contain 81 papers. The topics discussed include: a new approach to concurrent ring: 1 bit latency;image processing on parallel machines: a protocol for managing global objects;representation and measurement of non-rigid egocentric motion: a parallel implementation;efficient implementation of an abstract programming model for image processing on transputers;distributed environments & distributed program design;parallel discrete event simulation using space-time events;deterministic parallel execution of sequential programs;a new massively parallel architecture relying on asynchronous communications;and creatures and spirals a data parallel object architecture.
In this paper we propose a scalable admission control scheme for the QoS sensitivity traffic in core-stateless networks. In our scheme, the ingress routers perform admissibility test in a fully distributed and paralle...
详细信息
ISBN:
(纸本)9780769535579
In this paper we propose a scalable admission control scheme for the QoS sensitivity traffic in core-stateless networks. In our scheme, the ingress routers perform admissibility test in a fully distributed and parallel fashion for requests by dint of our bandwidth dividing based virtual link mechanism. Then, we introduce a novel two phase token passing mechanism to adaptively optimize the proportion of bandwidth dividing among contending edge routers according to the dynamic of their traffic. In addition, we adopt a measurement based admission decision-making criterion to gain the benefit of high utilization of statistical multiplexing. Our simulation results are very promising indicating that even under very high request load it is possible to perform admission control and resource allocation in parallel without suffering in terms of response time, packet loss rate, or utilization.
The carrier-null message approach to conservative distributed discrete-event simulation can significantly reduce the number of synchronization messages required to avoid deadlock. In thts paper we show that the origin...
详细信息
ISBN:
(纸本)1565550277
The carrier-null message approach to conservative distributed discrete-event simulation can significantly reduce the number of synchronization messages required to avoid deadlock. In thts paper we show that the original approach does not apply to simulations with arbitrary communication graphs and we propose a modified carrier-null approach whtch does. We present and discuss some preliminary results obtained using our approach to simulate digital logic circuits.
We show that the latest version of massively parallelprocessing associative string processing architecture (System-V) is applicable for fast Monte Carlo simulation if an effective on-processor random number generator ...
详细信息
ISBN:
(纸本)0769500595
We show that the latest version of massively parallelprocessing associative string processing architecture (System-V) is applicable for fast Monte Carlo simulation if an effective on-processor random number generator is implemented. Our lagged Fibonacci generator can produce 10(8) random numbers on a processor string of 12K PE-s. The time dependent Monte Carlo algorithm of the one-dimensional non-equilibrium kinetic Ising model perfoms 80 faster than the corresponding serial algorithm on a 300 MHz UltraSparc.
暂无评论