In this paper we address the problem of efficiently performing parallel discrete-event simulation in the case that events elaboration is dependent of other processes' local states. We propose a parallelsimulation...
详细信息
ISBN:
(纸本)0818679654
In this paper we address the problem of efficiently performing parallel discrete-event simulation in the case that events elaboration is dependent of other processes' local states. We propose a parallelsimulation policy, called State Query Time Warp (SQTW), based on the Time Warp mechanism. We present experiments performed by means of a SQTW-based parallel simulator on a T-800 transputer machine for solving performance models based on state-dependent routing queueing network models. The experiments are used for assessing overheads and efficiency involved by SQTW;results show that high efficiency is achievable, and surprisingly reveal SQTW is able to globally reduce rollback overheads with respect to corresponding Time Warp simulations.
This paper introduces a technique to control the overhead of time management processes in order to make such mechanisms appropriate for real-time distributedsimulation. A novel message accounting scheme, the offset-e...
详细信息
ISBN:
(纸本)0769519709
This paper introduces a technique to control the overhead of time management processes in order to make such mechanisms appropriate for real-time distributedsimulation. A novel message accounting scheme, the offset-epoch method, is presented as a way to increase the efficiency of time management algorithms by eliminating transient messages. A synchronized lower-bound on timestamp (LBTS) computation exploits this efficiency to reduce time management overheads. This approach represents one step in bridging the gap that now exists between analytic and real-time distributedsimulations.
Time-parallelsimulation defines a methodology that can be applied to certain specific simulation problems. In this paper, we present a time-parallel approach for trace-driven simulation of the CSMA/CD protocol. The &...
详细信息
ISBN:
(纸本)0769519709
Time-parallelsimulation defines a methodology that can be applied to certain specific simulation problems. In this paper, we present a time-parallel approach for trace-driven simulation of the CSMA/CD protocol. The "memoryless" property of the physical system under moderate traffic loads allows for efficient time-parallelsimulation. We also present two optimization techniques: the estimation of idle points and the incremental fix-up computation. The former can improve the probability that a subtrace begins with a known initial system state. The latter can speedup the fix-up computation that is required when the estimation of the initial state is incorrect.
The use of Web caches makes it possible to offer a service of better quality to Internet users. Due to the difficulty in obtaining a real network of caches to experience with, most research projects use Web caches sim...
详细信息
ISBN:
(纸本)0769513484
The use of Web caches makes it possible to offer a service of better quality to Internet users. Due to the difficulty in obtaining a real network of caches to experience with, most research projects use Web caches simulators. This paper presents a Web cache simulator capable of executing simulations of distributed and cooperative Web caches servers. ne simulator allows the measurement of the impact of configuration parameters in the cache performance, making possible to design better cooperative cache networks. To show an example of the simulator use, simulations based in the Brazilian National Research Network - RNP are presented.
Large-scale ecological simulations are natural candidates for distributed discrete event simulation. In optimistic simulation of spatially explicit models, a difficult problem arises when individuals migrate between p...
详细信息
ISBN:
(纸本)0818679654
Large-scale ecological simulations are natural candidates for distributed discrete event simulation. In optimistic simulation of spatially explicit models, a difficult problem arises when individuals migrate between physical regions simulated by different logical processes. We present a solution to this problem that uses shared object states. Shared states allow for efficient communication between LPs and for early detection of canceled events. We briefly describe an optimistic simulation environment called EcoKit, which operates on top of the WarpKit implementation of Time Warp. Our experiments with this system on a shared memory multiprocessor show that EcoKit promises to scale well both with the number of processors and the number of individuals simulated.
Data distribution management (DDM) is concerned with the problem of efficiently distributing state information among the entities in a distributedsimulation. While heavily studied in the DIS community for training ap...
详细信息
Data distribution management (DDM) is concerned with the problem of efficiently distributing state information among the entities in a distributedsimulation. While heavily studied in the DIS community for training applications using real-time synchronization, this problem has received considerably less attention for logical time simulations, and little performance data has appeared in this regard. This paper is concerned with evaluating the performance of a logical time synchronized.
This paper advocates the use of a formal framework for analyzing simulation performance. simulation performance is characterized based on the three simulation development process boundaries: physical system, simulatio...
详细信息
ISBN:
(纸本)0769521118
This paper advocates the use of a formal framework for analyzing simulation performance. simulation performance is characterized based on the three simulation development process boundaries: physical system, simulation model, and simulator implementation. Firstly, we formalize simulation event ordering using partially ordered set theory. A simulator implements a simulation event ordering, and incurs implementation overheads when enforcing event ordering at runtime. Secondly, we apply our formalism to extract and formalize the simulation event orderings of both sequential and parallelsimulations. Thirdly, we propose the relation stricter and a measure called strictness for comparing and quantifying the degree of event dependency of simulation event orderings respectively.
Great effort has been devoted to the design of optimized checkpointing strategies for optimistic parallel discrete event simulators. On the other hand there is less work in the direction to improve the execution mode ...
详细信息
ISBN:
(纸本)076951104X;0769511058
Great effort has been devoted to the design of optimized checkpointing strategies for optimistic parallel discrete event simulators. On the other hand there is less work in the direction to improve the execution mode of any single checkpoint operation. Specifically, checkpoint operations are typically charged to the CPU, thus leading to freezing of the simulation application while checkpointing is in progress, i.e. the execution mode of the checkpointing protocol is typically synchronous. In this paper we focus on improvements of the execution mode and present a software architecture, designed for myrinet based Network of Workstations (NOWs), to avoid application freezing during any checkpoint operation, thus moving the execution itself towards an asynchronous mode. This is done by charging checkpoint operations to a hardware component distinct from the CPU, namely a DMA engine. On the other hand, totally asynchronous checkpointing could suffer from data inconsistency whenever the content of a state buffer is accessed for further modifications while a checkpoint operation involving it is not yet completed. To avoid this, the architecture includes functionalities for resynchronization on demand. We have used these functionalities to implement an execution mode of the checkpointing protocol we refer to as semi-asynchronous. By the results of all experimental study we argue that the semi-asynchronous mode can be an effective solution to almost completely remove the delay associated with any checkpoint operation from the completion time of the simulation.
A probabilistic scheduling policy for selecting the next event to be processed by a processor during optimistic parallel discrete event simulation is presented. This policy exploits the knowledge about statistical dis...
详细信息
A probabilistic scheduling policy for selecting the next event to be processed by a processor during optimistic parallel discrete event simulation is presented. This policy exploits the knowledge about statistical distributions of the difference between the timestamps of successively committed events of a logical process. simulation results are presented to demonstrate that under appropriate conditions probabilistic scheduling generates substantially fewer number of rollbacks compared to the lowest timestamp first scheduling policy.
暂无评论