The paper presents a new architecture for systems based on run-time reconfigured shared memory processor clusters meant for implementation using network on chip technology. Clusters constitute local data exchange sub-...
详细信息
ISBN:
(纸本)0769522106
The paper presents a new architecture for systems based on run-time reconfigured shared memory processor clusters meant for implementation using network on chip technology. Clusters constitute local data exchange sub-networks, which dynamically connect processors with shared memory modules. The sub-networks enable exposure of data from one processor's data cache for reading by other processors to their data caches. This inter-processor data exchange paradigm, called "communication on the fly", enables direct communication between processor data caches. Dual-ported data caches are assumed to enable parallel reading and writing data between the caches and memory modules. In the proposed architecture, programs are executed according to a cache-controlled macro data flow execution model. Computational tasks are so defined, as to eliminate re-loading of data caches during task execution. A special program macro-data flow graph representation enables modeling of program behaviour for different architectural and program structure assumptions. simulation results of symbolic execution of program graphs of matrix multiplication are presented in the paper. They show suitability of the proposed architecture for very fine grain parallel computations.
Large complex system simulation in various fields of science and engineering requires tremendous computational resources;however sequential execution algorithms badly limited its performance. So recently there has bee...
详细信息
ISBN:
(纸本)3540240764
Large complex system simulation in various fields of science and engineering requires tremendous computational resources;however sequential execution algorithms badly limited its performance. So recently there has been a great deal of interest in parallel and distributedsimulation, which runs on multiple processors to accelerate simulation. This paper begins with introduction of synchronization mechanisms. The emphasis of this paper is to provide and describe the implementation of the flexible cycle algorithm. This improved algorithm solves some fatal problems of conservative or optimistic algorithms, resulting in the best of both methods. Finally we also analyze how to compute the performance parameter M of this algorithm in detail.
The paper deals with simulation and analysis tools of control system with distributed inputs and outputs based on TCP/IP and UDP/IP protocols. These protocols are not strictly prepared for industrial control applicati...
详细信息
The proceedings contain 190 papers. The special focus in this conference is on Grid Application, Peer to Peer Computing and Grid Architectures. The topics include: A computer systems approach to grid research;problems...
ISBN:
(纸本)3540219889
The proceedings contain 190 papers. The special focus in this conference is on Grid Application, Peer to Peer Computing and Grid Architectures. The topics include: A computer systems approach to grid research;problems of and mechanisms for instantiating virtual organizations;making grid computing real for high performance and enterprise computing;scientific issues, methodology, and practice in china;enabling scientific study of dynamic grid behavior;on-demand business collaboration enablement with services computing;a research on the framework of grid manufacturing;implementation of grid-enabled medical simulation applications using workflow techniques;a new overlay network based on can and chord;a distributed media service system based on Globus data-management technologies;a web-GIS based urgent medical rescue CSCW system for SARS disease prevention;a model for adapting service-oriented grid applications;conceptual framework for recommendation system based on distributed user ratings;the design and implementation of the gridlab information service;a new navigation method for web users;application availability measurement in computational grid;research and application of distributed fusion system based on grid computing;an efficient and self-configurable publish-subscribe system;the implementation of the genetic optimized algorithm of air craft geometry designing based on grid computing;the design of adaptive platform for visual-intensive applications over the grid;some grid automata for grid computing;the granularity analysis of MPI parallel programs;integration of the distributedsimulation into the OGSA model and an extendable grid simulation environment based on gridsim.
The proceedings contain 26 papers. The special focus in this conference is on Membrane Computing. The topics include: Proton pumping P systems;a binary data structure for membrane processors;parsing with active P auto...
ISBN:
(纸本)354020895X
The proceedings contain 26 papers. The special focus in this conference is on Membrane Computing. The topics include: Proton pumping P systems;a binary data structure for membrane processors;parsing with active P automata;universality of minimal symport/antiport;collapsing hierarchies of parallel rewriting P systems without target conflicts;evolution and observation;tiling rectangular pictures with P systems;simulating boolean circuits with P systems;P systems running on a cluster of computers;implementing in prolog an effective cellular solution to the knapsack problem;a petri net view;P systems generating hexagonal picture languages;a membrane system for the leukocyte selective recruitment;P systems with cutting/recombination rules assigned to membranes;the number of membranes matters;an agent-based behavioural model of monomorium pharaonis colonies;can hyperbolic geometry be of help for P systems;a linear-time solution to the knapsack problem using P systems with active membranes;a reconfigurable hardware membrane system;P systems and petri nets;simulation of mobile ambients by P systems;computing partial recursive functions by transition P systems;P systems with external input and learning strategies;a distributedsimulation of transition P systems and about splicing P systems with immediate communication and non-extended splicing P systems.
The proceedings contain 42 papers. The special focus in this conference is on Experimental and Efficient Algorithms. The topics include: A hybrid bin-packing heuristic to multiprocessor scheduling;efficient edge-swapp...
ISBN:
(纸本)3540220674
The proceedings contain 42 papers. The special focus in this conference is on Experimental and Efficient Algorithms. The topics include: A hybrid bin-packing heuristic to multiprocessor scheduling;efficient edge-swapping heuristics for finding minimum fundamental cycle bases;solving chance-constrained programs combining tabu search and simulation;an algorithm to identify clusters of solutions in multimodal optimisation;on an experimental algorithm for revenue management for cargo airlines;cooperation between branch and bound and evolutionary approaches to solve a bi-objective flow shop problem;simple max-cut for split-indifference graphs and graphs with few P4’s;a randomized heuristic for scene recognition by graph matching;an efficient implementation of a joint generation algorithm;lempel, even, and cederbaum planarity method;a greedy approximation algorithm for the uniform labeling problem analyzed by a primal-dual technique;distributed circle formation for anonymous oblivious robots;dynamic programming and column generation based approaches for two-dimensional guillotine cutting problems;engineering shortest path algorithms;how to tell a good neighborhood from a bad one;implementing approximation algorithms for the single-source unsplittable flow problem;fingered multidimensional search trees;faster deterministic and randomized algorithms on the homogeneous set sandwich problem;efficient implementation of the BSP/CGM parallel vertex cover FPT algorithm and combining speed-up techniques for shortest-path computations.
This paper presents wrapping techniques that we have developed in order to integrate applications into the scientific workflow management system TENT. These techniques take into account the varying degrees of controll...
详细信息
This paper presents wrapping techniques that we have developed in order to integrate applications into the scientific workflow management system TENT. These techniques take into account the varying degrees of controllability found for each application and utilize various kinds of communication mechanisms between application and the associated wrapper. We also give an outline of the principles of component technology employed in the CORBA based TENT environment. In addition, the means by which applications can be connected through their associated wrappers and TENT components to form distributed workflows are presented. It is demonstrated how TENT provides the user with a homogeneous perspective on scientific workflows running in a distributed environment and how he is enabled to efficiently configure and control these workflows by means of the TENT component model.
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.
暂无评论