Mathematical optimization algorithms are ubiquitous in computational science and engineering where the objective function of the optimization problem involves a complicated computer model predicting relevant phenomena...
详细信息
ISBN:
(纸本)9789897580369
Mathematical optimization algorithms are ubiquitous in computational science and engineering where the objective function of the optimization problem involves a complicated computer model predicting relevant phenomena of a scientific or engineering system of interest. therefore, in this area of mathematical software, it is indispensable to combine software for optimization with software for simulation, typically developed independently of each other by members of separate scientific communities. From a software engineering point of view, the situation becomes even more challenging when the simulation software is developed using a parallel programming paradigm without taking into consideration that it will be executed within an optimization context. the EFCOSS environment alleviates some of the problems by serving as an interfacing layer between optimization software and simulation software. In this paper, we show the software design of those parts of EFCOSS that are relevant to the integration of a simulation software involving different parallel programming paradigms. the parallel programming paradigms supported by EFCOSS include MPI for distributed memory and OpenMP for shared memory. In addition, the simulation software can be executed on a remote parallel computer.
parallelcomputing is increasingly used to provide more performance to applicationsthat need tremendous computational power. the main characteristics of distributedparallel machines are heterogeneity, dynamism and s...
详细信息
ISBN:
(纸本)9781424425785
parallelcomputing is increasingly used to provide more performance to applicationsthat need tremendous computational power. the main characteristics of distributedparallel machines are heterogeneity, dynamism and size. they influence directly the way the application and platform monitoring tasks are performed, especially when analyzing a large quantity of information collected in a topologically complex machine. this paper describes our efforts to provide parallel programmers and Grid users a new way to visualize monitoring data. Using graphics in three dimensions and information visualization techniques, we aim at bringing rich topological information to the rendered scene. It results in an immersible and human readable representation of complex monitoring data, suited to Grid environments. We first review known techniques in information visualization context, especially those that address the case of hierarchical information, and we discuss about their use in our context. then, we propose a new 3D approach that combines the classical space-time visualization of application traces withthe representation of the application's communication pattern. Finally, we present experimental results obtained through the visualization of parallelapplications in our prototype.
Traditionally, checkpointing techniques have been used to secure the execution of sequential and parallel programs. this article shows that checkpointing techniques can also be used to automatically generate a paralle...
详细信息
ISBN:
(纸本)9783540739395
Traditionally, checkpointing techniques have been used to secure the execution of sequential and parallel programs. this article shows that checkpointing techniques can also be used to automatically generate a parallel program from a sequential program, this program being executed on any kind of distributedparallel system. the article also presents how this new technique have been included inside the usual compilation chain to provide a distributed implementation of OpenMP. Finally, some performance measurements are discussed.
Withthe continuous development of GPUs, modern general-purpose computation on GPUs (GPGPUs) is providing growing parallelism to general programs besides graphics applications. However, for those programs that involve...
详细信息
ISBN:
(纸本)9780769548791
Withthe continuous development of GPUs, modern general-purpose computation on GPUs (GPGPUs) is providing growing parallelism to general programs besides graphics applications. However, for those programs that involve both CPU and GPU, the data transmission bandwidth between them may become bottleneck that prevents GPU from fully exploiting its parallelcomputing capacity. As to avoid the defect, we try to reduce the data transmission by keeping part of the computation tasks on the CPU side other than sending all the data over to the GPU and process there. In this way the computation is done on CPU and GPU in parallel, and therefore also reduces overall process time. In order to split the computation workload in a systematic approach, we try to divide the corresponding data into chunks of proper size. We experimented our data dividing and heterogeneous memory scheduling with 2 benchmarks. the matrix multiplication is more than 30% faster, and the k - means2D is nearly 10% faster, than running solely in GPU.
the paper presents the pricing strategies for automatic agent, which is design for e-commerce in the reality. Given the characters of e-commerce in reality, the paper firstly model the bargaining process;Secondly clas...
详细信息
ISBN:
(纸本)0769524052
the paper presents the pricing strategies for automatic agent, which is design for e-commerce in the reality. Given the characters of e-commerce in reality, the paper firstly model the bargaining process;Secondly classify the bargaining scenarios and present the optimal bargaining strategy under sub-incomplete information for corresponding bargaining scenarios;then extend them to the optimal pricing strategies under incomplete information condition;thirdly, discuss the conditions for convergence of optimal strategy;Finally, the analysis shows that the bargaining strategic profiles form the sequential equilibrium and the agreement is unique under certain conditions.
Context-aware computing is a computing paradigm in which appiications can discover and take advantage of context information. Faced withthe diverse mix of distributed and heterogeneous sensors, embedded devices and m...
详细信息
ISBN:
(纸本)0769524052
Context-aware computing is a computing paradigm in which appiications can discover and take advantage of context information. Faced withthe diverse mix of distributed and heterogeneous sensors, embedded devices and mobile devices in the ubiquitous computing environments, it is fairly difficult for users to develop context-aware applications. this paper proposes ScvdWare, a middleware to support the development of context-aware applications. We introduce the developing language, compiling and assembling of a context-aware application. To illustrate the developing process, we design an example which shows how the application adapts to changes in the environments.
A parallel version of a new automatic Harris-based corner detector is presented. A scheduler to dynamically and homogeneously distribute high computational workload on heterogeneous parallel architectures such as Grid...
详细信息
ISBN:
(纸本)9781424452910
A parallel version of a new automatic Harris-based corner detector is presented. A scheduler to dynamically and homogeneously distribute high computational workload on heterogeneous parallel architectures such as Grid systems has been implemented to speedup the whole procedure. Experimental results show the robustness of the underlying scheduler, which can be easily exploited in various automatic image analysis systems.
parallel Discrete Event Simulation (PDES) allows to speed up the execution of a simulation by distributing the simulation9;s workload between multiple processors. the mapping of the simulation entities over the pro...
详细信息
ISBN:
(纸本)9789549641523
parallel Discrete Event Simulation (PDES) allows to speed up the execution of a simulation by distributing the simulation's workload between multiple processors. the mapping of the simulation entities over the processors is very important with regard to the efficiency of the overall process of simulation. this paper presents an approach for automation of the process of mapping over distributed environment (network of workstations) for optimistic simulation based on the synchronization protocol Time Warp.
the proceedings contain 232 papers. the topics discussed include: convolutional neural network-based approach for citrus diseases recognition;using temporal conceptual graphs and neural networks for big data-based att...
ISBN:
(纸本)9781728143286
the proceedings contain 232 papers. the topics discussed include: convolutional neural network-based approach for citrus diseases recognition;using temporal conceptual graphs and neural networks for big data-based attack scenarios reconstruction;taxi demand prediction with LSTM-based combination model;counting attention based on classification confidence for visual question answering;an optimization method of WebP images lossy compression algorithm for FPGAs;self-adaptive address mapping mechanism for access pattern awareness on DRAM;a high-performance self-learning antelopes migration algorithm (SAMA) for global optimization;and using resource use data and system logs for HPC system error propagation and recovery diagnosis.
暂无评论