A short presentation and a benchmark of a prototyping tool that facilitates the use of a network of computers or a parallel computer to solve course-grained large-scale problems are the subjects of this paper. the ben...
详细信息
ISBN:
(纸本)3540437924
A short presentation and a benchmark of a prototyping tool that facilitates the use of a network of computers or a parallel computer to solve course-grained large-scale problems are the subjects of this paper. the benchmark is concerning systems in a range of tens to hundreds ordinary differential equations solved by parallel numerical methods.
Currently existing monitoring services for Grid infrastructures typically collect information from local agents and store it as data sets in global repositories However, for some scenarios querying real-time streams o...
详细信息
ISBN:
(纸本)9783642144028
Currently existing monitoring services for Grid infrastructures typically collect information from local agents and store it as data sets in global repositories However, for some scenarios querying real-time streams of monitoring information would be extremely useful In this paper, we evaluate Complex Event processing technologies applied to real-time Grid monitoring. We present a monitoring system which uses CEP technologies to expose monitoring information as queryable data streams We study an example use case - monitoring for job rescheduling We also employ CEP technologies for data reduction, measure the overhead of monitoring, and conclude that teal-tune Grid monitoring is possible without excessive intrusiveness for resources and network
Cloud Computing has long extended beyond the original focus of providing scalable on-demand resources for web applications and is now also ubiquitous in batch-style data processing applications. Employing Cloud servic...
详细信息
ISBN:
(纸本)9781467394697
Cloud Computing has long extended beyond the original focus of providing scalable on-demand resources for web applications and is now also ubiquitous in batch-style data processing applications. Employing Cloud services for data analysis tasks is also a viable alternative for researchers who are limited by their locally available compute power and in the need for timely execution. However, the provisioning and deployment of machines and applications at Infrastructure-as-a-Service (IaaS) providers is a non-trivial task for the average scientist. Within this paper, we propose a framework for automating the deployment and execution of existing applications in a data-parallel fashion in Cloud environments with only negligible effort by the user. Our evaluation of a real-world scientific use case exhibits a significant speedup compared to local execution.
We investigate parallel algorithms for the solution of the shallow-water equation in a space-time framework. For periodic solutions, the discretized problem can be written as a large cyclic non-linear system of equati...
详细信息
ISBN:
(纸本)9783642314995;9783642315008
We investigate parallel algorithms for the solution of the shallow-water equation in a space-time framework. For periodic solutions, the discretized problem can be written as a large cyclic non-linear system of equations. this system of equations is solved with a Newton iteration which uses two levels of preconditioned GMRES solvers. the parallel performance of this algorithm is illustrated on a number of numerical experiments.
Frequent items in high-speed streaming data are important to many applications like network monitoring and anomaly detecting. To deal with high arrival rate of streaming data, it is desirable that such systems be capa...
详细信息
ISBN:
(纸本)9781509007684
Frequent items in high-speed streaming data are important to many applications like network monitoring and anomaly detecting. To deal with high arrival rate of streaming data, it is desirable that such systems be capable of supporting high processingthroughput with tight guarantees on errors. In this paper, we address the problem of finding frequent and top-k items, and present a parallel version of the Space Saving algorithm in the context of the open source distributed computing system. Based on the theoretical analysis, the errors are restrictively bounded in our algorithm, and our parallel design could achieve high throughput. Taking advantage of the distributed computing resources, our evaluation reveals that such design delivers linear speedup with remarkable scalability.
We reconsider the familiar problem of executing a perfectly parallel workload consisting of N independent tasks on a parallel computer with P << N processors. We show that there are memory-bound problems for whi...
详细信息
ISBN:
(纸本)9783642552243
We reconsider the familiar problem of executing a perfectly parallel workload consisting of N independent tasks on a parallel computer with P << N processors. We show that there are memory-bound problems for which the runtime can be reduced by the forced parallelization of individual tasks across a small number of cores. Specific examples include solving differential equations, performing sparse matrix-vector multiplications, and sorting integer keys.
the modeling of the electrical activity of the heart is of great medical and scientific interest, because it provides a way to get a better understanding of the related biophysical phenomena, allows the development of...
详细信息
ISBN:
(纸本)9783642143892
the modeling of the electrical activity of the heart is of great medical and scientific interest, because it provides a way to get a better understanding of the related biophysical phenomena, allows the development of new techniques for diagnoses and serves as a platform for drug tests. the cardiac electrophysiology may be simulated by solving a partial differential equation (PDE) coupled to a system of ordinary differential equations (ODEs) describing the electrical behavior of the cell membrane. the numerical solution is, however, computationally demanding because of the fine temporal and spatial sampling required. the demand for real time high definition 3D graphics made the new graphic processing units (GPUs) a highly parallel, multithreaded, many-core processor with tremendous computational horsepower. It makes the use of GPUs a promising alternative to simulate the electrical activity in the heart. the aim of this work is to study the performance of the use of GPUs to solve the equations underlying the electrical activity in a simple cardiac tissue.
the goal of this paper is to propose a methodology of the effective cost function determination for the job shop scheduling problem in parallel computing environment. parallel Random Access Machine (PRAM) model is app...
详细信息
ISBN:
(纸本)9783642314995;9783642315008
the goal of this paper is to propose a methodology of the effective cost function determination for the job shop scheduling problem in parallel computing environment. parallel Random Access Machine (PRAM) model is applied for the theoretical analysis of algorithm efficiency. the methods need a fine-grained parallelization, therefore the approach proposed is especially devoted to parallel computing systems with fast shared memory. the methods proposed are tested with CUDA and OpenCL and ran on NVidia and ATI GPUs.
Neural networks and deep learning currently provide the promising solutions to many practical problems. One of the difficulties in building neural network models is the training process that requires to find an optima...
详细信息
ISBN:
(纸本)9781450353168
Neural networks and deep learning currently provide the promising solutions to many practical problems. One of the difficulties in building neural network models is the training process that requires to find an optimal solution for the network weights. the Particle Swarm Optimization (PSO) algorithm has been recently applied to neural network training due to its global search. However, the PSO algorithm suffers from large execution time. In this paper, a parallel design of the PSO algorithm is proposed, using OpenCL language on a GPU. To improve the performance, fine memory allocation is considered for the parallel particle processing and an efficient parallel reduction scheme based on local and global reduction is proposed. By fully utilizing the processing power of the GPU, the OpenCL PSO implementation accelerates the neural network training by up to 35 times, compared to the multithreaded C++ implementation on a CPU.
Subspace methods have been applied in various application fields to obtain robust results. Using multilinear algebra, they can also be applied on structured tensorial data. this work combines this principle withthe p...
详细信息
ISBN:
(纸本)9789897580697
Subspace methods have been applied in various application fields to obtain robust results. Using multilinear algebra, they can also be applied on structured tensorial data. this work combines this principle withthe power of non-linear kernels to investigate its merits in single trial classification for a mobile BCI ERP classification task. the accuracy difference with regard to more conventional vector kernels is evaluated for sitting and walking condition, increasing training data set and averaging over multiple trials. the study concludes that in general, the tensorial approach does not yield any advantage, though it might for specific subjects.
暂无评论