This article presents a scheduling algorithm that assigns tasks represented in a directed acyclic graph (DAG). The behavior of the proposed algorithm is compared with the dominant sequence clustering (DSC) algorithm o...
详细信息
This article presents a scheduling algorithm that assigns tasks represented in a directed acyclic graph (DAG). The behavior of the proposed algorithm is compared with the dominant sequence clustering (DSC) algorithm on a set of DAG tests. The tests were carried out on two sets of DAGs, one with tasks of arbitrary duration, and the other with tasks of unit time duration. When comparing the results for each algorithm in terms of parallel computation time, the proposed algorithm is superior in 75% of the cases, equal in 20% of the tests, and in those cases in which DSC outperforms the proposed algorithm (5% of the cases), it can be seen that the planning carried out by DSC uses a greater number of processors than the proposed algorithm, so that the performance in terms of CPU use is lower for DSC. The results obtained show that the proposed algorithm is superior in 100% of the cases when considering the number of processors used.
The proceedings contain 20 papers. The topics discussed include: HLA-based adaptive distributedsimulation of wireless mobile systems;implementation of federation management services over federation community networks...
ISBN:
(纸本)0769519709
The proceedings contain 20 papers. The topics discussed include: HLA-based adaptive distributedsimulation of wireless mobile systems;implementation of federation management services over federation community networks;parallel network simulation under distributed genesis;enabling large-scale multicast simulations by reducing memory requirements;an implementation of the SSF scalable simulation framework on the Cray MTA;Predictable Time Management for Real-Time distributedsimulation;scalable RTI-based parallelsimulation of networks;reducing the size of routing tables for large-scale network simulation;utility analysis of parallelsimulation;a macroscopic behavior model for self-timed pipeline systems;hybrid packet/fluid flow network simulation;large-scale TCP models using optimistic parallelsimulation;and Maya: a multi-paradigm network modeling framework for emulating distributed applications.
Large-scale parallel discrete event simulations of massive networks, such as the Internet, are "Grand Challenge" problems: packet level simulation of even a small fraction of the Internet would consume the r...
详细信息
ISBN:
(纸本)0769519709
Large-scale parallel discrete event simulations of massive networks, such as the Internet, are "Grand Challenge" problems: packet level simulation of even a small fraction of the Internet would consume the resources of the most powerful computers available. We reimplement the SSF Scalable simulation Framework so we can run large-scale network simulations originally written for DaSSF Our implementation, CraySSF is designed for the Cray-MTA, a multi-threaded supercomputer architecture developed specifically to address large-scale computations of the kind that are not easily distributed. This paper describes the architecture, implementation issues, and preliminary performance results on a variety of (stock) serial and parallel architectures.
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.
There is a wide-spread usage of hardware design languages(HDL) to speed up the time-to-market for the design of modern digital systems. Verification engineers can simulate hardware in order to verify its performance a...
详细信息
ISBN:
(纸本)0769519709
There is a wide-spread usage of hardware design languages(HDL) to speed up the time-to-market for the design of modern digital systems. Verification engineers can simulate hardware in order to verify its performance and correctness with help of an HDL. However simulation can't keep pace with the growth in size and complexity of circuits and has become a bottleneck of the design process. distributed HDL simulation on a cluster of workstations has the potential to provide a cost-effective solution to this problem. In this paper we describe the design and implementation of DVS, an object-oriented framework for distributed Verilog simulation. Verilog is an HDL which sees wide industrial use. DVS is an outgrowth of Clustered Time Warp, originally developed for logic simulation. The design of the framework emphasizes simplicity and extensibility and aims to accommodate experiments involving partitioning and dynamic load balancing. Preliminary results obtained by simulating a 16bit multiplier are presented.
We present a classification that groups lookback into four types: direct strong lookback, universal strong lookback, direct weak lookback, and universal weak lookback. They are defined in terms of absolute and dynamic...
详细信息
ISBN:
(纸本)0769519709
We present a classification that groups lookback into four types: direct strong lookback, universal strong lookback, direct weak lookback, and universal weak lookback. They are defined in terms of absolute and dynamic impact times. We discuss relationships between lookback types by considering,when rollbacks and/or anti-messages are avoided From different types of lookback, we also derive three optimization techniques for optimistic simulation and point out their advantages over lazy cancellation. Finally, we show that all four types of lookback exist in the PCS network simulation and can be exploited by either lookback-based or optimistic protocols.
We describe two major developments in the General Network simulation Integration System (Genesis): the support for BGP protocol in large network simulations and distribution of the simulation memory among Genesis comp...
详细信息
ISBN:
(纸本)0769519709
We describe two major developments in the General Network simulation Integration System (Genesis): the support for BGP protocol in large network simulations and distribution of the simulation memory among Genesis component simulations. Genesis uses a high granularity synchronization mechanism between parallelsimulations simulating parts of a network. This mechanism uses checkpointed simulation state to iterate over the same time interval until convergence. It also replaces individual packet data for flows crossing the network partitions with statistical characterization of such flows over the synchronization time interval. We had achieved significant performance improvement over the sequential simulation for simulations with TCP and UDP traffic. However, this approach can not be used directly to simulate dynamic routing protocols that use underlying network for exchanging protocol information, as no packets are exchanged in Genesis between simulated network parts. We have developed a new mechanism to exchange and synchronize BGP routing data among distributed Genesis simulators. The extended Genesis allows simulations of more realistic network scenarios, including routing flows, in addition to TCP or UDP data traffic. Large memory size required by simulation software hinders the simulation of large-scale networks. Based on our new support of distributed BGP simulation, we developed an approach to construct and simulate networks on distributed memory using Genesis simulators in such a way that each participating processor possesses only data related to the part of the network it simulates. This solution supports simulations of large-scale networks on machines with modest memory size.
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.
暂无评论