Optimistic techniques can improve the performance of discrete-event simulations, but one area where optimistic simulators have been unable to show performance improvement is in the simulation of parallel programs. Unf...
详细信息
ISBN:
(纸本)0769511058
Optimistic techniques can improve the performance of discrete-event simulations, but one area where optimistic simulators have been unable to show performance improvement is in the simulation of parallel programs. Unfortunately, parallel program simulation using direct execution is difficult;the use of direct execution implies that the memory and computation requirements of the simulator are at least as large as that of the target application, which restricts the target systems and application problem sizes that can be studied. Memory usage is especially important for optimistic simulators due to the need for periodic state-saving and rollback. In our research we addressed this problem and have implemented a simulation library running a Time-Warp-based optimistic engine that uses direct execution to simulate and predict the performance of parallel MPI programs while attaining good simulation speedup. For programs with data sets too large to be directly executed with our optimistic simulator, we reduced the memory and computational needs of these programs by utilizing a static task graph and code-slicing methodology, an approach which also exhibited good performance speedup.
Real-time distributedsimulations, such as on-line gaming or military training simulations are normally considered to be non-deterministic. Analysis of these simulations is therefore difficult depending solely on logg...
详细信息
ISBN:
(纸本)0769506771
Real-time distributedsimulations, such as on-line gaming or military training simulations are normally considered to be non-deterministic. Analysis of these simulations is therefore difficult depending solely on logging and runtime observations. This paper explores art approach for removing one major source of nondeterminism in these simulations, thereby allowing repeatable executions. Specifically, we use a synchronization protocol to ensure repeatable delivery of messages. Through limited instrumentation of the simulation code, we maintain a virtual time clock, by which message delivery! is governed. The additional overhead imposed by the scheme is shown to be reasonable, although additional reductions in this overhead are anticipated The results are demonstrated in the context of a simple combat model whose only source of non-determinism is communications latency. The simulation is shown to be tirade repeatable, and the perturbation on the execution compared to the nonrepeatable execution small. The paper is one step in bridging the gap between the traditional PDES perspective and real-time simulation world.
In this paper, we present two new methods to simulate Petri Nets: a data parallelsimulation and a distributedsimulation. Both simulations use an equational representation of the net in the so called (min, +) algebra...
详细信息
In this paper, we present two new methods to simulate Petri Nets: a data parallelsimulation and a distributedsimulation. Both simulations use an equational representation of the net in the so called (min, +) algebra. The data parallelsimulation is based on the use of matrix representation of these equations, and the distributedsimulation on the decomposition of a Petri Net into marked graph components.
With fixed lookahead information in a simulation model, the overhead of asynchronous conservative parallelsimulation lies in the mechanism used for propagating time updates in order for logical processes to safety ad...
详细信息
ISBN:
(纸本)076951104X;0769511058
With fixed lookahead information in a simulation model, the overhead of asynchronous conservative parallelsimulation lies in the mechanism used for propagating time updates in order for logical processes to safety advance their local simulation clocks. Studies have shown that a good scheduling algorithm should preferentially schedule processes containing events on the critical path. This paper introduces a lock-free algorithm for scheduling logical processes in conservative parallel discrete-event simulation on shared-memory multiprocessor machines. The algorithm uses fetch&add operations that help avoid inefficiencies associated with using locks. The lock-free algorithm is robust. Experiments show that, compared with the scheduling algorithm using locks, the lock-free algorithm exhibits better performance when the number of logical processes assigned to each processor is small or when the workload becomes significant. In models with large number of logical processes, our algorithm shows only modest increase in execution time due to the overhead in the algorithm for extra bookkeeping.
The simulation of wide area computer networks is one area where the benefits of parallelsimulation have been clearly demonstrated Here we present a description of a system that uses a parallel discrete event simulato...
详细信息
ISBN:
(纸本)0769506771
The simulation of wide area computer networks is one area where the benefits of parallelsimulation have been clearly demonstrated Here we present a description of a system that uses a parallel discrete event simulator to act as a high speed network emulator With this, real Internet Protocol (IP) traffic generated by application programs running on user workstations can interact with modelled traffic in the emulator thus providing a controlled test environment for distributed applications. The network emulator uses the TasKit conservative parallel discrete event simulation (PDES) kernel. TasKit has been shown to be able to achieve improved parallel performance over existing conservative and optimistic PDES kernels, as well as improved sequential performance over an existing central-event-list based kernel. This paper explains the modifications that have been made to TasKit to enable real-time operation along with the emulator interface that allows the IP network simulation running in the TasKit kernel to interact with real IP clients. Initial emulator performance data is included.
If a model shall be executed in a parallel, distributed instead of a sequential manner typically the entire simulation engine has to be exchanged. To adapt the simulation layer more easily to the requirements of a con...
详细信息
ISBN:
(纸本)0769521118
If a model shall be executed in a parallel, distributed instead of a sequential manner typically the entire simulation engine has to be exchanged. To adapt the simulation layer more easily to the requirements of a concrete model to be run in a specific environment a component based simulation layer has been developed for JAMES. A set of different simulator components demonstrates that a component-based design facilitates the exchange of simulators and their combination.
Strong reasons exist for executing a large-scale discrete-event simulation on a cluster of processor nodes (each of which may be a shared-memory multiprocessor or a uniprocessor). This is the architecture of the large...
详细信息
ISBN:
(纸本)076951104X;0769511058
Strong reasons exist for executing a large-scale discrete-event simulation on a cluster of processor nodes (each of which may be a shared-memory multiprocessor or a uniprocessor). This is the architecture of the largest scale parallel machines, and so the largest simulation problems can only be solved this way. It is a common architecture even in less esoteric settings, and is suitable for memory-bound simulations. This paper describes our approach to porting the SSF simulation kernel to this architecture, using the Message Passing Interface (MPI) system. The notable feature of this transformation is to support an efficient two-level synchronization and communication scheme that addresses cost discrepancies between shared-memory and distributed memory. In the initial implementation, we use a globally synchronous approach between distributed-memory noes, and an asynchronous shared-memory approach within a SMP cluster. The SSF API reflects inherently shared-memory assumptions;we report therefore on our approach for porting an SSF kernel to a cluster of SMP nodes. Experimental results on two architectures are described, for a model of TCP/IP traffic flows over a hierarchical network. The performance on a distributed network of commodity SMPs connected through ethernet is seen to frequently exceed performance on a Sun shared-memory multiprocessor.
A novel micro-kernel approach to building parallel/distributedsimulation systems is presented. Using this approach, a unified system architecture is developed for incorporating, multiple types of simulation processes...
详细信息
ISBN:
(纸本)0769523838
A novel micro-kernel approach to building parallel/distributedsimulation systems is presented. Using this approach, a unified system architecture is developed for incorporating, multiple types of simulation processes. The processes hold potential to employ a variety of synchronization mechanisms, and could even alter their choice of mechanism dynamically. Supported mechanisms include traditional lookahead-based conservative and state saving-based optimistic execution approaches. Also supported are newer mechanisms such as reverse computation-based optimistic execution and aggregation-based event processing, all within a single parsimonious application programming interface. The internal implementation and a preliminary performance evaluation of this interface are presented in mu sik, which is an efficient parallel/distributed realization of the microkernel architecture in C++.
parallel Monte Carlo simulation requires reliable RNGs. For sequential machines, good generators exist. A is not at all trivial to find high-quality RNGs for parallel machines. In this paper we present a short review ...
详细信息
ISBN:
(纸本)0818684577
parallel Monte Carlo simulation requires reliable RNGs. For sequential machines, good generators exist. A is not at all trivial to find high-quality RNGs for parallel machines. In this paper we present a short review of the main concepts to produce random numbers on parallel processors and, further, we illustrate some phenomena that occur with parallelization.
In real-time simulation, the system being simulated should display the same timing behavior as the target system. The simulation accuracy is increased as the simulation time unit is decreased. Although there are sever...
详细信息
ISBN:
(纸本)0769515762
In real-time simulation, the system being simulated should display the same timing behavior as the target system. The simulation accuracy is increased as the simulation time unit is decreased. Although there are several models for such a system, the TMO model is particularly appropriate due to its natural support for real-time distributed object oriented programming. This paper discusses the results of the implementation of a real-time airplane-landing simulator on a distributed computing environment using the TMO model.
暂无评论