scholarly journals Data Parallel Skeletons in Java

2012 ◽  
Vol 9 ◽  
pp. 1817-1826 ◽  
Author(s):  
Herbert Kuchen ◽  
Steffen Ernsting
Keyword(s):  
Data Parallel ◽  
Parallel Skeletons

Foundations of Data-parallel Skeletons

2003 ◽  
pp. 1-27 ◽  
Author(s):  
Jörg Fischer ◽  
Sergei Gorlatch ◽  
Holger Bischof
Keyword(s):  
Data Parallel ◽  
Parallel Skeletons

scholarly journals THE INTEGRATION OF TASK AND DATA PARALLEL SKELETONS

2002 ◽  
Vol 12 (02) ◽  
pp. 141-155 ◽  
Author(s):  
HERBERT KUCHEN ◽  
MURRAY COLE
Keyword(s):  
Parallel Programming ◽  
Travelling Salesman Problem ◽  
Travelling Salesman ◽  
Data Parallel ◽  
Programming Library ◽  
Parallel Skeletons

We describe a skeletal parallel programming library which integrates task and data parallel constructs within an API for C++. Traditional skeletal requirements for higher orderness and polymorphism are achieved through exploitation of operator overloading and templates, while the underlying parallelism is provided by MPI. We present a case study describing two algorithms for the travelling salesman problem.

ON IMPLEMENTING THE FARM SKELETON

2008 ◽  
Vol 18 (01) ◽  
pp. 117-131 ◽  
Author(s):  
MICHAEL POLDNER ◽  
HERBERT KUCHEN
Keyword(s):  
Load Balancing ◽  
Parallel Programming ◽  
Message Passing ◽  
Experimental Results ◽  
Algorithmic Skeletons ◽  
Data Parallel ◽  
Advantages And Disadvantages ◽  
Parallel Skeletons ◽  
Task Parallel ◽  
Communication Modes

Algorithmic skeletons intend to simplify parallel programming by providing a higher level of abstraction compared to the usual message passing. Task and data parallel skeletons can be distinguished. In the present paper, we will consider several approaches to implement one of the most classical task parallel skeletons, namely the farm, and compare them w.r.t. scalability, overhead, potential bottlenecks, and load balancing. We will also investigate several communication modes for the implementation of skeletons. Based on experimental results, the advantages and disadvantages of the different approaches are shown. Moreover, we will show how to terminate the system of processes properly.

Data-parallel line relaxation method for the Navier-Stokes equations

AIAA Journal ◽  
1998 ◽  
Vol 36 ◽  
pp. 1603-1609 ◽  
Author(s):  
Michael J. Wright ◽  
Graham V. Candler ◽  
Deepak Bose
Keyword(s):  
Stokes Equations ◽  
Relaxation Method ◽  
Parallel Line ◽  
Navier Stokes ◽  
Navier Stokes Equations ◽  
Data Parallel

scholarly journals Multi-GPU approach to global induction of classification trees for large-scale data mining

2021 ◽  
Author(s):  
Krzysztof Jurczuk ◽  
Marcin Czajkowski ◽  
Marek Kretowski
Keyword(s):  
Data Mining ◽  
Large Scale ◽  
Real Life ◽  
Population Based ◽  
Tree Structure ◽  
Global Approach ◽  
Data Parallel ◽  
Large Scale Data ◽  
The Impact ◽  
Scale Data

AbstractThis paper concerns the evolutionary induction of decision trees (DT) for large-scale data. Such a global approach is one of the alternatives to the top-down inducers. It searches for the tree structure and tests simultaneously and thus gives improvements in the prediction and size of resulting classifiers in many situations. However, it is the population-based and iterative approach that can be too computationally demanding to apply for big data mining directly. The paper demonstrates that this barrier can be overcome by smart distributed/parallel processing. Moreover, we ask the question whether the global approach can truly compete with the greedy systems for large-scale data. For this purpose, we propose a novel multi-GPU approach. It incorporates the knowledge of global DT induction and evolutionary algorithm parallelization together with efficient utilization of memory and computing GPU’s resources. The searches for the tree structure and tests are performed simultaneously on a CPU, while the fitness calculations are delegated to GPUs. Data-parallel decomposition strategy and CUDA framework are applied. Experimental validation is performed on both artificial and real-life datasets. In both cases, the obtained acceleration is very satisfactory. The solution is able to process even billions of instances in a few hours on a single workstation equipped with 4 GPUs. The impact of data characteristics (size and dimension) on convergence and speedup of the evolutionary search is also shown. When the number of GPUs grows, nearly linear scalability is observed what suggests that data size boundaries for evolutionary DT mining are fading.

scholarly journals Data Parallel Hypersweeps for in Situ Topological Analysis

2020 ◽  
Author(s):  
Petar Hristov ◽  
Gunther H. Weber ◽  
Hamish A. Carr ◽  
Oliver Rubel ◽  
James P. Ahrens
Keyword(s):  
Topological Analysis ◽  
Data Parallel
Sign in / Sign up

Export Citation Format

Share Document