Presents a study of scheduling dynamically evolving parallel programs in distributed multiprocessor systems. Four centralized schedulers based on a genetic algorithm are proposed. These schedulers consider both load b...
详细信息
The proceedings contain 86 papers. The special focus in this conference is on Workshops, and Several Associated Events. The topics include: A problem solving environment based on commodity software;a virtual programmi...
ISBN:
(纸本)9783540675532
The proceedings contain 86 papers. The special focus in this conference is on Workshops, and Several Associated Events. The topics include: A problem solving environment based on commodity software;a virtual programming environment for high performance parallel computing;grid computing on the web using the globus toolkit;algorithms for generic tools in parallel numerical simulation;dynamic grid adaption for computational magnetohydrodynamics;parallelization of irregular problems based on hierarchical domain representation;dynamic iterative method for fast network partitioning;a family of parallel incomplete cholesky preconditioners;a parallel block preconditioner accelerated by coarse grid correction;towards an execution system for distributed business processes in a virtual enterprise;towards a multi-layer architecture for scientific virtual laboratories;modelling control systems in an event-driven coordination language;ruling agent motion in structured environments;dynamic reconfiguration in coordination languages;adding flexibility in a cooperative workflow execution engine;a web-based distributed programming environment;interoperability support in distributed on-line monitoring systems;run-time optimization using dynamic performance prediction;performance portability for skeletal programming;a novel distributed algorithm for high-throughput and scalable gossiping;run-time support to register allocation for loop parallelization of image processing programs;a java-based parallel programming support environment;task farm computations in java;simulating job scheduling for clusters of workstations;effective communication support for parallel applications running on clusters of commodity workstation;distributedparallel query processing on networks of workstations and explicit schemes applied to aeroacoustic simulations.
This paper examines implementations of a multi-layer perceptron (MLP) on bus-based shared memory (SM) and on distributed memory (DM) multiprocessor systems. The goal has been to optimize HW and SW architectures in ord...
详细信息
This paper examines implementations of a multi-layer perceptron (MLP) on bus-based shared memory (SM) and on distributed memory (DM) multiprocessor systems. The goal has been to optimize HW and SW architectures in order to obtain the fastest response possible. Prototyping parallel MLP algorithms for up to 8 processing nodes with the DM as well as SM memory was done using CSP-based TRANSIM tool. The results of prototyping MLPs of different sizes on various number of processing nodes demonstrate the feasible speedups, efficiency and time responses for the given CPU speed, link speed or bus bandwidth.
The Automated Guided Vehicles (or AGVs for short) have become an important option in the container handling process. We present a parallel simulation system for the study of AGV routing schemes. We model the AGV syste...
详细信息
The Automated Guided Vehicles (or AGVs for short) have become an important option in the container handling process. We present a parallel simulation system for the study of AGV routing schemes. We model the AGV system based on a time-driven approach and execute the model on an efficient simulation engine implemented by using Cilk, a parallel programming language developed at MIT. We will elaborate on the AGV routing algorithms and traffic control scheme and propose a deadlock-free decentralized routing scheme. The performance results of the scheme are also documented.
This paper describes the characteristics of an optimal routing that assigns each arriving packet to one of two heterogeneous parallel servers. The characteristics are derived from numerical solutions to a routing prob...
详细信息
This paper describes the characteristics of an optimal routing that assigns each arriving packet to one of two heterogeneous parallel servers. The characteristics are derived from numerical solutions to a routing problem, which finds an optimal routing that minimizes the average packet delay under the condition that the input traffic is completely deterministic. Four characteristics are presented. (1) Under light or moderate traffic, the average packet delay of the optimal routing is almost the same as that of the join the shortest delay (JSD) policy. (2) As traffic becomes heavier, the optimal routing more often uses the fix queue based on size (FS) policy. (3) Under heavy traffic, the optimal routing assigns small packets to a slower server. (4) As the ratio of the two servers' service rates C1/C2 (C1 &le C2) decreases, under lighter traffic the optimal routing uses the FS policy. A new adaptive routing policy based on these four characteristics outperforms the JSD policy.
This paper describes an infrastructure that enables transparent development of image processing software for parallel computers. The infrastructure's main component is an image processing library containing operat...
详细信息
ISBN:
(纸本)0769507506;0769507514
This paper describes an infrastructure that enables transparent development of image processing software for parallel computers. The infrastructure's main component is an image processing library containing operations capable of running on distributed memory MIMD-style parallel hardware. Since the library has a programming interface identical to that of a familiar sequential image processing library, the parallelism is completely hidden from the user. All library functions are based on an abstract parallel image processing machine (APIPM), introduced in this paper. To guide the process of automatic parallelization and optimization, a performance model is defined for operations implemented using APIPM instructions. Experiments show that for realistic image operations performance predictions are highly accurate. These results suggest the the infrastructure's core forms a powerful basis for automatic parallelization and optimization of complete applications.
One of the major problems in an electrical insulation structure is its failure while operating at normal voltage stress which is due to electrical treeing. It is well-known that this electrical tree grows progressivel...
详细信息
The proceedings contain 189 papers. The special focus in this conference is on Support Tools, Evironments, Performance Evaluation, Prediction and Load Balancing. The topics include: Grid-based asynchronous migration o...
ISBN:
(纸本)9783540679561
The proceedings contain 189 papers. The special focus in this conference is on Support Tools, Evironments, Performance Evaluation, Prediction and Load Balancing. The topics include: Grid-based asynchronous migration of execution context in java virtual machines;visualization and computational steering in heterogeneous computing environments;a web-based finite element meshes partitioner and load balancer;an infrastructure for the construction of memory hierarchy analysis tools;on combining computational differentiation and toolkits for parallel scientific computing;generating parallel program frameworks from parallel design patterns;a callgraph-based search strategy for automated performance diagnosis;automatic performance analysis of MPI applications based on event traces;an extensible environment for visualizing multi-threaded programs executions;a statistical-empirical hybrid approach to hierarchical memory analysis;use of performance technology for the management of distributedsystems;delay behavior in domain decomposition applications;automating performance analysis from UML design patterns;integrating automatic techniques in a performance analysis session;the specific cluster operation and performance evaluation benchmark suite;a fast and accurate approach to analyze cache memory behavior;performance prediction of a NAS benchmark program with chronosmix environment;a hierarchical approach to irregular problems;the impact of migration on parallel job scheduling for distributedsystems;exploiting knowledge of temporal behaviour in parallel programs for improving distributed mapping;preemptive task scheduling for distributedsystems and scheduling trees with large communication delays on two identical processors.
The purpose of this study is to demonstrate the capabilities of distributed-parallel computing methods to model previously intractable multi-body dynamic systems, and apply them to structural analysis. Generally, stru...
详细信息
Genetic Algorithms are becoming a common tool for optimal design applications, where, due to the multiple solutions issue, global search techniques are required Anyway, when dealing with real problems, involving sever...
详细信息
ISBN:
(纸本)076950759X
Genetic Algorithms are becoming a common tool for optimal design applications, where, due to the multiple solutions issue, global search techniques are required Anyway, when dealing with real problems, involving several degrees of freedom, the actual computing power restricts the global search ability. The availability of cheap hardware has recently caused the spreading of multiprocessors computing systems. In particular new genetic techniques have been proposed to adapt the-method's characteristics to the parallel architecture allowing in this way also to deal with real problems.:One of these techniques, called niching approach, can be implemented by dividing the population into subgroups, and letting each group to evolve on one of the processors, interacting only when scheduled. Objective of this work is to discuss the perspectives of niching approaches in the electromagnetic optimal design applications. As an example case, preliminary results about SMES (Superconducting Magnetic Energy Storage) devices are proposed.
暂无评论