Skip to content
Free planNo credit card requiredStart writing
AI for Research home
AI for Research home
Dashboard
Sign inGet started

Search 250 million research papers

Read the abstract, see who wrote it and who cites it, and get the citation in APA, MLA, Chicago or Harvard.

About 10,000 papers for “Parallel processing”

Cluster-Based Parallel Image Processing

Jeffrey M. Squyres, Andrew Lumsdaine, Brian C. Mccandless et al. · 1996

Many image processing tasks exhibit a high degree of data locality and parallelism and map quite readily to specialized massively parallel computing hardware. However, as workstation clusters are becoming a viable and economical parallel computing resource, it is important to understand how to use these environments for parallel image processing as well. In this paper we discuss our implementation of a parallel image processing software library (the Parallel Image Processing Toolkit). The library is easily extensible and hides most parallelism from the user. Inside the Toolkit, a message-passing model of parallelism is designed around the Message Passing Interface (MPI) standard. Experimental results are presented to demonstrate the parallel speedup obtained with the Parallel Image Processing Toolkit in a typical workstation cluster with some common image processing tasks. We also discuss load balancing and the potential for parallelizing portions of image processing tasks that seem to...

Cited by 0

Parallel processing of XML databases

Ghassan Z. Qadah · 2006

Beowulf cluster is a name given to a high performance, low-cost parallel computer system made of commodity hardware and software components. It consists of a number of processing nodes, interconnected via a switch. The extensible markup language (XML) data model, on the other hand, has recently gained huge popularity because of its ability to represent a wide variety of structured (tabular-like) and semi-structured (textual-like) data. Several query languages have been proposed for the XML data model, the most-widely known is XQuery. This paper reviews the XML data model and its query language within the context of cluster/parallel computing environment. It examines several techniques for structuring and storing XML data across the different cluster nodes. It develops a number of algorithms suitable for processing a certain class of queries, namely, the containment queries, against the parallel XML database. This paper also shows that one of these algorithms, the one that takes advantage of the parallelism existing between the different documents within the XML database, is outperforming all of the other presented ones

Cited by 8

Guest Editors' Introduction Parallel Processing

H.J. Segel, Leah H. Jamieson · IEEE Transactions on Computers · 1984

AS we approach the limits of how much a single processor's speed can be increased through advances in technology, there is no question that parallelism will be the key to building tomorrow's fastest, mostpowerful computers. Current parallel processing studies demonstrate that parallelism has the potential to provide solutions to a wide range of computationally intensive problems. Nonetheless, at a time when "supercomputer" is almost a household word, fundamental issues in parallel processing continue to demand attention. This Special Issue is evidence of the scope of the field of parallel processing. Topics addressed include synchronous and asynchronous architectures; communications via buses, trees, pipelines, crossbars, and multistage inter-connection networks; algorithms for numerical matrix operations, image generation (with Star Trek II cited as an example), dictionary searches, and database queries; efficient scheduling of hundreds of tasks; automatic vectorization of constructs that are not directly vectorizable; and performance evaluations of all of the above. A brief introduction to the papers follows.

Cited by 1

The limits of parallel processing

Katrin Göthe · publish.UP (University of Potsdam) · 2009

Trying to do two things at once decreases performance of one or both tasks in many cases compared to the situation when one performs each task by itself. The present thesis deals with the question why and in which cases these dual-task costs emerge and moreover, whether there are cases in which people are able to process two cognitive tasks at the same time without costs. In four experiments the influence of stimulus-response (S-R) compatibility, S-R modality pairings, interindividual differences, and practice on parallel processing ability of two tasks are examined. Results show that parallel processing is possible. Nevertheless, dual-task costs emerge when: the personal processing strategy is serial, the two tasks have not been practiced together, S-R compatibility of both tasks is low (e.g. when a left target has to be responded with a right key press and in the other task an auditorily presented “A” has to be responded by saying “B”), and modality pairings of both tasks are Non Standard (i.e., visual-spatial stimuli are responded vocally whereas auditory-verbal stimuli are responded manually). Results are explained with respect to executive-based (S-R compatibility) and content-based crosstalk (S-R modality pairings) between tasks. Finally, an alternative information processing account with respect to the central stage of response selection (i.e., the translation of the stimulus to the response) is presented.

Cited by 1Open access

Parallel processing and expert systems

Jerry Yan, Sonie Lau · NASA STI Repository (National Aeronautics and Space Administration) · 1991

Whether it be monitoring the thermal subsystem of Space Station Freedom, or controlling the navigation of the autonomous rover on Mars, NASA missions in the 1990s cannot enjoy an increased level of autonomy without the efficient implementation of expert systems. Merely increasing the computational speed of uniprocessors may not be able to guarantee that real-time demands are met for larger systems. Speedup via parallel processing must be pursued alongside the optimization of sequential implementations. Prototypes of parallel expert systems have been built at universities and industrial laboratories in the U.S. and Japan. The state-of-the-art research in progress related to parallel execution of expert systems is surveyed. The survey discusses multiprocessors for expert systems, parallel languages for symbolic computations, and mapping expert systems to multiprocessors. Results to date indicate that the parallelism achieved for these systems is small. The main reasons are (1) the body of knowledge applicable in any given situation and the amount of computation executed by each rule firing are small, (2) dividing the problem solving process into relatively independent partitions is difficult, and (3) implementation decisions that enable expert systems to be incrementally refined hamper compile-time optimization. In order to obtain greater speedups, data parallelism and application parallelism must be exploited.

Cited by 1Open access

The Future of Parallel Processing

Kenneth J. Thurber, Peter C. Patton · IEEE Transactions on Computers · 1973

This correspondence considers the cost and performance factors in computer architecture and how they apply to parallel processors.

Cited by 10

Realities associated with parallel processing

Daniel Platt, Andrea Rossi, Jenny Wells et al. · 1989

At the T. J. Watson Research Center, there is a very active Condensed Matter Physics Group engaged in the study of semiconductors such as silicon (Si) and gallium-arsenide (Ga-As)1. One of the most important computer codes developed at Watson is a Density Functional Program which is used to study the electronic structure of semiconductors. This program also consumes the most CPU time of all other production applications at Watson. Thus, it was decided to undertake the parallelization of this code not only with the hope of reducing elapsed time, improving turnaround on the IBM 30902, and conserving other system resources such as memory and disk space, but also in an attempt to test the IBM Parallel FORTRAN Compiler3 on a production program while developing an understanding of the impact of parallel jobs from a system standpoint.

Cited by 0Open access

Introduction: Parallel and Distributed Processing

IEEE Transactions on Computers · 1982

THE GROWTH of interest in parallel and distributed processing in the last decade has been explosive. It will doubtlessly continue unabated, being the only approach to increasing the speed and power of computing systems whose ultimate limits, by fundamental physical laws or well-established theory, are not yet definable. Some of the papers that follow in this Special Issue on Parallel and Distributed Processing were originally presented at the 1981 (Tenth) International Conference on Parallel Procesing; others were submitted in response to the call for papers for the Special Issue. Being devoted to recent advances in parallel and distributed processing systems, the Special Issue covers a wide range of subjects, from architectures to algorithms, from protocols to performance evaluation. They, and other topics, may eventually justify special issues of their own.

Cited by 0

Computer graphics and parallel processing

John Lansdown · ITNOW · 1987

Abstract One of the most active specialist groups in the BCS is the Displays Group. Under the leadership of its Secretary, the indefatigable Rae Earnshaw and his committee, the group holds regular and well-attended one-day conferences on all sorts of topics related to computer graphics and display technology. One of these recently covered issues in Parallel Processing for Displays and was chaired by Tom Heywood of IBM. John Lansdown comments on what might be learned from the eight papers presented at that meeting.

Cited by 0

Parallel processing of genomics data

Giuseppe Agapito, Pietro Hiram Guzzi, Mario Cannataro · AIP conference proceedings · 2016

The availability of high-throughput experimental platforms for the analysis of biological samples, such as mass spectrometry, microarrays and Next Generation Sequencing, have made possible to analyze a whole genome in a single experiment. Such platforms produce an enormous volume of data per single experiment, thus the analysis of this enormous flow of data poses several challenges in term of data storage, preprocessing, and analysis. To face those issues, efficient, possibly parallel, bioinformatics software needs to be used to preprocess and analyze data, for instance to highlight genetic variation associated with complex diseases. In this paper we present a parallel algorithm for the parallel preprocessing and statistical analysis of genomics data, able to face high dimension of data and resulting in good response time. The proposed system is able to find statistically significant biological markers able to discriminate classes of patients that respond to drugs in different ways. Experiments performed on real and synthetic genomic datasets show good speed-up and scalability.

Cited by 3

Parallel strategies for SAR processing

Jesus A. Segoviano · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2004

This article proposes a series of strategies for improving the computer process of the Synthetic Aperture Radar (SAR) signal treatment, following the three usual lines of action to speed up the execution of any computer program. On the one hand, it is studied the optimization of both, the data structures and the application architecture used on it. On the other hand it is considered a hardware improvement. For the former, they are studied both, the usually employed SAR process data structures, proposing the use of parallel ones and the way the parallelization of the algorithms employed on the process is implemented. Besides, the parallel application architecture classifies processes between fine/coarse grain. These are assigned to individual processors or separated in a division among processors, all of them in their corresponding architectures. For the latter, it is studied the hardware employed on the computer parallel process used in the SAR handling. The improvement here refers to several kinds of platforms in which the SAR process is implemented, shared memory multicomputers, and distributed memory multiprocessors. A comparison between them gives us some guidelines to follow in order to get a maximum throughput with a minimum latency and a maximum effectiveness with a minimum cost, all together with a limited complexness. It is concluded and described, that the approach consisting of the processing of the algorithms in a GNU/Linux environment, together with a Beowulf cluster platform offers, under certain conditions, the best compromise between performance and cost, and promises the major development in the future for the Synthetic Aperture Radar computer power thirsty applications in the next years.

Cited by 0