the authors explore parallel processing issues, such as relationships between speedup and problem partitioning schemes, and problem size and time duration of each iteration for PIC (particle in cell) method on differe...
详细信息
ISBN:
(纸本)0818621443
the authors explore parallel processing issues, such as relationships between speedup and problem partitioning schemes, and problem size and time duration of each iteration for PIC (particle in cell) method on different multiprocessors. A partitioning scheme, hybrid partitioning, is introduced. Hybrid partitioning has evolved out of two general approaches to PIC problem decomposition on multiprocessors, partitioning particles and partitioning the space. the authors chose the shared memory multiprocessor environment for analyzing the parallel (distributed computing) algorithms. Two different BBN Butterfly machines (GP1000 and TC2000) were employed as testbeds.
作者:
Ilic, MarijaKosanic, MiroslavMIT
EECS Dept 77 Massachusetts Ave Cambridge MA 02139 USA MIT
LIDS 77 Massachusetts Ave Cambridge MA 02139 USA
In this paper we exploit the structure of dynamic models for general cyber-physical systems (CPS). Key to this structure is a unifying notion of an interaction variable between components/subsystems and the neighbouri...
详细信息
ISBN:
(纸本)9798350336825
In this paper we exploit the structure of dynamic models for general cyber-physical systems (CPS). Key to this structure is a unifying notion of an interaction variable between components/subsystems and the neighbouring sub-systems within a multi-layered CPS. the higher-level aggregate model derived in terms of unifying energy and power dynamics explicitly captures dynamic interactions of interest. In this paper this structure is utilised further to formulate a distributed interactive algorithm for simulating dynamic interactions and for aligning them according to generalised Tellegen's theorem. Proof of concept simulations are bench-marked against the centralised simulations of a simple DC circuit serving constant power load (CPL). Notably, the simulation results reflect the underlying physics and explicitly capture oscillations between neighbouring modules. this algorithm lends itself to multi-layered parallel implementation by means of minimal information exchange. As such, it supports simulating complex electric energy systems, in particular power-electronically controlled DC micro-grids, and, more generally multi-energy systems.
Excessive rollback recoveries due to overoptimistic event execution in Time Warp simulators often degrade their runtime performance. this paper presents a two-sided throttling scheme to dynamically adjust the event ex...
详细信息
Excessive rollback recoveries due to overoptimistic event execution in Time Warp simulators often degrade their runtime performance. this paper presents a two-sided throttling scheme to dynamically adjust the event execution speed of Time Warp simulators. the proposed throttle is based on a new concept called global progress window, which allows the individual simulation process to be positioned on a global time scale, thereby to accelerate or suspend their event execution. As each simulation process can be throttled to a steady state, excessive rollback recoveries due to causality errors can be avoided. To quantify the effect of rollbacks and for purpose of comparing different Time Warp implementations, we propose two new measures called RPE (number of Rollback events Per committed Event), and Ε (relative Effectiveness in reducing rollback overhead). Our implementation results show that the proposed throttle effectively regulates the proceeding of each simulation process, resulting in a significant reduction in rollback thrashing and elapsed time.
Determinism plays a key role in both model verification and in achieving accurate simulation statistical results. In this paper we present a general framework for achieving determinism in the execution Of hybrid simul...
详细信息
ISBN:
(纸本)9780769530116
Determinism plays a key role in both model verification and in achieving accurate simulation statistical results. In this paper we present a general framework for achieving determinism in the execution Of hybrid simulation models. this framework guarantees that models will produce the same results irrespective to the simulation kernel they are executed. thus, hybrid models can produce the same trace when executed in sequential or parallel machines and in this case when conservative or optimistic approaches are employed. For describing hybrid flow models we use the Heterogeneous Flow System Specification (HFSS), a hierarchical and modular formalism that can represent systems with a dynamic topology. the paper also solves the problems introduced by these systems in achieving deterministic simulations.
this paper presents the design and implementation of a distributed data-driven workflow system on top of the TeraGrid infrastructure. the workflow system is based on a data management architecture that provides easy a...
详细信息
ISBN:
(纸本)9780769528106
this paper presents the design and implementation of a distributed data-driven workflow system on top of the TeraGrid infrastructure. the workflow system is based on a data management architecture that provides easy access to scientific data collections via the TeraGrid network. the workflow system allows researchers to construct scientific workflows for data discovery, access, transformation, and analysis. the system leverages JOpera, an open-source workflow engine and visual composer as well as a set of web service-based data and computation modules. To demonstrate its effectiveness, we create an end-to-end climate simulation data analysis workflow that connects the data management architecture to YeraGrid computation resources. We also develop a workflow monitoring service to keep track of distributed workflow execution..
In this paper we describe web-based distributed system suitable for acceleration of fault simulation in digital circuits. Framework has three-tier client server concept. Java applets are used for user interfaces. Java...
详细信息
ISBN:
(纸本)9781424422760
In this paper we describe web-based distributed system suitable for acceleration of fault simulation in digital circuits. Framework has three-tier client server concept. Java applets are used for user interfaces. Java servlet on master server supports communication, task partitioning, user tracking, data management, etc. HTTP protocol based solution is used, since it is well established and flexible in firewall-protected environments. Reuse of existing test tools by encapsulation into Java agents extends the lifecycle and value of these tools. Considerable fault simulation speedup was gained in experiments. Two types of fault set partitioning were tried: adjacent fault selection and random fault selection. the latter is able to ensure more equal execution times for subtasks and therefore contributes to overall shorter parallelsimulation time.
Recently, the virtual reality has progressed rapidly in the field of the science and technology. the coverage includes extremely various fields such as communication, design, entertainment, etc. Various VR space have ...
详细信息
ISBN:
(纸本)0769522815
Recently, the virtual reality has progressed rapidly in the field of the science and technology. the coverage includes extremely various fields such as communication, design, entertainment, etc. Various VR space have been proposed and constructed by present. However, there are no VR system that realizes real time simulation based on the same physical law as it works in the real world. then, we construct a virtual air hockey game based on rigid dynamics in this research. Moreover, to provide the presence with users, two sensations (the visual and auditory sensation) are synchronized. this paper describes various problems and their solutions in construction of the system, and utilities for synchronizing the two sensations.
A possible solution to guarantee critical requirements in Web Services designs is the use of an autonomic architecture, able to auto-configure and to auto-tune. this paper presents an innovative approach for the devel...
详细信息
ISBN:
(纸本)0769522815
A possible solution to guarantee critical requirements in Web Services designs is the use of an autonomic architecture, able to auto-configure and to auto-tune. this paper presents an innovative approach for the development of self-optimizing autonomic systems for web services architectures, based on the adoption of a simulation engine for obtaining performance predictions. MAWeS (MetaPL/HeSSE Autonomic Web Services) is a framework whose aim is to support the development of self-optimizing predictive autonomic systems for Web service architectures. It adopts a simulation-based methodology, which allows to predict system performances in different status and load conditions. the predicted results are used for a feed forward control of the system, which self-tunes before the new conditions and the subsequent performance losses are actually observed.
Effective communication among agents in large teams is crucial because the members share a common goal but only have partial views of the environment. Information sharing is difficult in a large team because, a team m...
详细信息
ISBN:
(纸本)9780769528106
Effective communication among agents in large teams is crucial because the members share a common goal but only have partial views of the environment. Information sharing is difficult in a large team because, a team member may have a piece of valuable information but not know who needs the information, since it is infeasible to know what each other agent is doing. Information sharing is a main part of any system or organization. the information sharing needs to be fool proof Only the legitimate receiver should be able to get hold of the information. this paper mainly deals with intelligent software agents for information sharing with confidentiality and trust. It clearly defines an Intelligent Software Agent, background of Information sharing in intelligent agents and the trust in the agents. Some part of the information needs confidentiality. the information that is shared requires security policy enforced based on the domain of the information and trust level of individual agent. this paper also provides the results of a multi-agent simulation for sharing information. It also implements trust calculation based on the quality of information provided by the peer agents in the simulation.
the topology of a peer-to-peer overlay network is usually constructed randomly without considering the characteristics of underlying physical links. thus searching and routing between peer nodes is often inefficient. ...
详细信息
ISBN:
(纸本)0769522815
the topology of a peer-to-peer overlay network is usually constructed randomly without considering the characteristics of underlying physical links. thus searching and routing between peer nodes is often inefficient. this paper proposes a mechanism that a peer node autonomously optimizes the topology of a peer-to-peer network based on the metrics of underlying networks. the simulation results show that this mechanism optimizes the topology of a peer-to-peer network and significantly decreases the average hop countbetween any pair of peer nodes.
暂无评论