the process interaction world view is widely used in the general simulation community for its expressive power and is supported by most modem simulation languages. In parallel discrete event simulation, however its us...
详细信息
ISBN:
(纸本)0818679654
the process interaction world view is widely used in the general simulation community for its expressive power and is supported by most modem simulation languages. In parallel discrete event simulation, however its use remains comparatively rare due to the perceived inefficiency (and difficulty) of parallel implementations. We present a new implementation strategy for parallel process-oriented simulation languages. this innovative, semantics-based approach directly addresses two common concerns of such languages. By concentrating on the intrinsic threads of control, we avoid the proliferation of simulation objects (and their associated costs) that might result from a naive translation. More fundamentally, the primary costs associated with process-oriented languages - those of context switching between stacks and, in an optimistic setting, of saving the state of these stacks - are entirely eliminated since our explicit use of continuations avoids the need for stacks in the first place. We similarly obtain cheap and natural thread pre-emption.
Time Warp's optimistic scheduling requires the maintenance of simulation state history to support rollback in the event of causality violations. State history, and the ability to rollback the simulation, can provi...
详细信息
ISBN:
(纸本)0818679654
Time Warp's optimistic scheduling requires the maintenance of simulation state history to support rollback in the event of causality violations. State history, and the ability to rollback the simulation, can provide unique functionality for human-in-the-loop simulation environments. this paper investigates the use of Time Warp to output valid simulation state in a near real-time manner re-execute portions of the simulation, and interactively probe simulation values to ascertain underlying causes of transient behavior. A shared-memory, multi-threaded interactive simulation architecture is presented and the additional state saving requirements imposed by interactivity are examined. the shortcomings of existing state saving schemes lead us to propose Multiplexed State Saving (MSS). By interleaving checkpointing and incremental state logs MSS provides bounded rollback costs and asynchronous access to prior simulation state. the interaction algorithms and MSS form a scalable, bounded cost component suitable for use in a real-time interactive Time Warp system.
In this paper we revisit the notion of event simultaneity in the context of parallel and distributedsimulation. Although the simulation community has recognized this problem for years, it has focused mainly on the me...
详细信息
In this paper we revisit the notion of event simultaneity in the context of parallel and distributedsimulation. Although the simulation community has recognized this problem for years, it has focused mainly on the mechanics of breaking event-time ties and has neither measured its extent nor considered its implications. Extant simulators (both serial and parallel) prohibit simultaneity either by user-specified event priorities or by an arbitrary (but well-documented) tie-breaking mechanism. We shall show, theoretically and empirically, that these strategies may lead to an invalid simulation. In doing so, we shall introduce the threshold of event simultaneity and use it to understand the semantics of simultaneity.
the Conjugate Residual method, one of the iterative methods for solving linear systems, is applied to the problems with a dense coefficient matrix on distributed memory parallel-computers. Based on an assumption on th...
详细信息
the Conjugate Residual method, one of the iterative methods for solving linear systems, is applied to the problems with a dense coefficient matrix on distributed memory parallel-computers. Based on an assumption on the computation and communication times of the proposed algorithm for parallel computers, it is shown that the optimal number of processing elements is proportional to the problem size N. the validity of the prediction is confirmed through numerical experiments on Hitachi SR2201.
this paper presents the results of an experimental study to evaluate the effectiveness of multiple synchronization protocols and partitioning algorithms in reducing the execution time of switch-level models of VLSI ci...
详细信息
this paper presents the results of an experimental study to evaluate the effectiveness of multiple synchronization protocols and partitioning algorithms in reducing the execution time of switch-level models of VLSI circuits. Specific contributions of this paper include: (i) parallelizing an existing switch-level simulator such that the model can be executed using conservative and optimistic simulation protocols with minor changes, (ii) evaluating effectiveness of several partitioning algorithms for parallelsimulation, and (iii) demonstrating speedups with both consecutive and optimistic simulation protocols for seven circuits, ranging in size from 3 K transistors to about 87 K transistors.
We present an architecture for parallel computation of incompressible IISPH simulations on distributed memory systems. We use orthogonal recursive bisection for domain decomposition and present a stable and fast conve...
详细信息
We present in this paper a novel load balancing and rescheduling approach based on the concept of the Sandpile cellular automaton: a decentralized multi-agent system working in a critical state at the edge of chaos. O...
详细信息
We present in this paper a novel load balancing and rescheduling approach based on the concept of the Sandpile cellular automaton: a decentralized multi-agent system working in a critical state at the edge of chaos. Our goal is providing fairness between concurrent job submissions in highly parallel and distributed environments such as currently built cloud computing systems by minimizing slowdown of individual applications and dynamically rescheduling them to the best suited resources. the algorithm design is experimentally validated by a number of numerical experiments showing the effectiveness and scalability of the scheme in the presence of a large number of jobs and resources and its ability to react to dynamic changes in real time. (C) 2016 Elsevier B.V. All rights reserved.
the IDES project at Sandia National Laboratories is developing a large scale portable parallel simulator for use in stockpile stewardship. IDES will use the Breathing-Time-Buckets synchronization protocol;to support I...
详细信息
the IDES project at Sandia National Laboratories is developing a large scale portable parallel simulator for use in stockpile stewardship. IDES will use the Breathing-Time-Buckets synchronization protocol;to support IDES development, this paper studies a performance model and describes performance experiments on expected workload and architectural parameters. A new parallel algorithm for terminating the window quickly is also described and analyzed.
the efficiency of parallel Discrete Event simulations that use the optimistic protocol is strongly dependent on the overhead incurred by rollbacks. this paper introduces a novel approach to rollback processing which l...
详细信息
the efficiency of parallel Discrete Event simulations that use the optimistic protocol is strongly dependent on the overhead incurred by rollbacks. this paper introduces a novel approach to rollback processing which limits the number of events rolled back as a result of a straggler or antimessage. the method, called Breadth-First Rollback (BFR), is suitable for spatially explicit problems where the space is discretized and distributed among processes and simulation objects move freely in the space. BFR uses incremental state saving, allowing the recovery of causal relationships between events during rollback. these relationships are then used to determine which events need to be rolled back. Our results demonstrate an almost linear speedup - a dramatic improvement over the traditional approach to rollback processing.
the problem of executing sequential programs in parallel using the optimistic algorithm Time Warp is considered. this is done by first mapping the sequential execution to a control tree and then assigning timestamps t...
详细信息
the problem of executing sequential programs in parallel using the optimistic algorithm Time Warp is considered. this is done by first mapping the sequential execution to a control tree and then assigning timestamps to each node in the tree. For such timestamps to be effective in either hardware or software they must be finite, this implies that they must be periodically rescaled to allow old timestamps to be reused. A number of timestamp representations are described and compared on the basis of their complexity;the frequency and cost of rescaling;and the cost of performing basic operations, including comparison and creation of new timestamps.
暂无评论