Filters








6,975 Hits in 4.8 sec

Parallelization of quantifier elimination on a workstation network [chapter]

Hoon Hong
1993 Lecture Notes in Computer Science  
This paper reports our e ort to parallelize on a network of workstations the quanti er elimination algorithm over the reals which was devised by Collins and improved by the author.  ...  The preliminary experiments show both sub-linear and super-linear speedups due to speculative parallelism. On the problems we have tested so far, the e ciencies range between 70% and 320%.  ...  I thank George Collins and Bruno Buchberger for making various suggestions on the draft of this paper.  ... 
doi:10.1007/3-540-56686-4_42 fatcat:e6rttnjwtnfy5fxvzzftx5wrwq

Approaches to parallel quantifier elimination

Andreas Dolzmann, Oliver Gloor, Thomas Sturm
1998 Proceedings of the 1998 international symposium on Symbolic and algebraic computation - ISSAC '98  
We study parallel algorithms based on these methods for several parallelization environments: a Cray YMP4 T3D, a workstation cluster, and a multi-processor Sparc.  ...  Special-purpose quanti er elimination procedures for problems of low degree using virtual substitution of test terms have recently turned out to be applicable to a variety of non-trivial non-academic problems  ...  Acknowledgments We w ould like to thank Andreas Weber for many v aluable discussions, and Winfried Neun and the Konrad Zuse Center in Berlin for the cooperation which led to the parallelization on the  ... 
doi:10.1145/281508.281564 dblp:conf/issac/DolzmannGS98 fatcat:42xkuxrcjrb6nfkgn6hagfxbfi

Parallel search for LTL violations

Michael D. Jones, Jacob Sorber
2003 International Journal on Software Tools for Technology Transfer (STTT)  
A unique advantage of BEE is that it is well-suited for use on clusters of non-dedicated workstations.  ...  However, the capacity of parallel model checkers is in turn limited by available aggregate memory and network bandwidth.  ...  Because these workstations are networked and rarely fully utilized, they can be used to form a dynamic cluster of non-dedicated workstations.  ... 
doi:10.1007/s10009-003-0115-8 fatcat:7m4htyi4nbbw7ctljrqhtofbvi

Comparison of resource platform selection approaches for scientific workflows

Yogesh Simmhan, Lavanya Ramakrishnan
2010 Proceedings of the 19th ACM International Symposium on High Performance Distributed Computing - HPDC '10  
We use only very limited high-level information on the workflow length, width, and data sizes. The length and width are indicative of the workflow duration and parallelism.  ...  We believe the simplicity of the blackbox model can help inform a scientist on the applicability of cloud computing resources even before porting an existing workflow.  ...  Resource usage features of a workflow quantify the computational, data storage, and networking resources. The compute usage can be specified as time taken to run the stage on a specific core speed.  ... 
doi:10.1145/1851476.1851541 dblp:conf/hpdc/SimmhanR10 fatcat:x4oirqjyhfbvngeyc7br6d4hsm

Synchronization hardware for networks of workstations

Rahmat S. Hyder, David A. Wood
1996 Proceedings of the 10th international conference on Supercomputing - ICS '96  
Networks of workstations (NOWs) are gaining popularity as lower-cost alternatives to massively-parallel processors (MPPs) because of their ability to leverage high-performance commodity workstations and  ...  At these higher latencies and typical prices, a 32-node NOW with an addon synchronization network is cost effective for 10 of the 12 benchmarks, compared to a uniprocessor with the same memory capacity  ...  PAPERS [9] is a low-cost synchronization network which supports fine-grain execution on workstation clusters, specifically operations on data aggregates as in a data-parallel pro-gram and VLIW-style  ... 
doi:10.1145/237578.237610 dblp:conf/ics/HyderW96 fatcat:jw5jmsu3fbg55mxkt5z46ncu2a

A cost analysis of film image management and four PACS based on different network protocols

David Beard, Deniese Parrish, Dan Stevenson
1990 Journal of digital imaging  
Based on the assumptions outlined, highspeed-network PACS aro leas costly than those based on Iow-speed networks for all scenarios considered.  ...  But st what point in the futuro will PACS really cost leas thon film-based image management for a given hospital size; and how aro these costa affected by the choice of the digital communication network  ...  Ultranet is a hub oriented gigabit per second local arca network (LAN) technology based on wide parallel datapath bus architecture.  ... 
doi:10.1007/bf03170570 pmid:2092809 fatcat:sos7h6whwzgajpbdsqas2qhuoa

Cluster Computing [chapter]

Mark Baker, John Brooke, Ken Hawick, Rajkumar Buyya
2001 Lecture Notes in Computer Science  
This ownership often brings problems when attempting to form a cluster of workstations.  ...  Typically, there are three types of "owner": • Ones who use their workstations for sending and receiving mail or preparing papers, such as administrative staff, librarian, theoreticians, etc. • Ones involved  ...  PRM enables users to run sequential or parallel jobs on a network of workstations.  ... 
doi:10.1007/3-540-44681-8_100 fatcat:cr6rpynstjgufacwbmciciriha

Cluster computing: the commodity supercomputer

Mark Baker, Rajkumar Buyya
1999 Software, Practice & Experience  
The availability of high-speed networks and increasingly powerful commodity microprocessors is making the usage of clusters, or networks, of computers an appealing vehicle for cost effective parallel computing  ...  Finally, we summarise our findings and draw a number of conclusions relating to the usefulness and likely future of cluster computing.  ...  Parallel computing on a dynamically changing set of non-dedicated workstations is called adaptive parallel computing.  ... 
doi:10.1002/(sici)1097-024x(199905)29:6<551::aid-spe248>3.0.co;2-c fatcat:sxxtvtczcrb75fxl4olnbljiqi

An Integrated Simulation Environment for Parallel and Distributed System Prototying

Alan D. George, Ryan B. Fogarty, Jeff S. Markwell, Michael D. Miars
1999 Simulation (San Diego, Calif.)  
processing algorithms, architectures, and systems using a networked cluster of workstations.  ...  On the software side, it is challenging to map and tune a parallel program to run well on such a sophisticated machine.  ...  Acknowledgments The support provided by the Office of Naval Research on grant N00014-98-1-0188 is acknowledged and appreciated, as is the SCI equipment support from Dolphin Interconnect Solutions, Inc.  ... 
doi:10.1177/003754979907200502 fatcat:nic2u6qu4rghbkktzbr5hqkvky

Multilevel Performance Evaluation of Nvidia Grid VCA Using Iray and V-Ray Rendering Engines in 3DS Max Design

Kenneth Ritter III
2018 Internet of Things and Cloud Computing  
This study outlines the performance between a single-user, traditional high-performance workstation and a multi-user, virtualized workstation.  ...  One is a diminishing return in GPU power output that is observed after allocating four or more GPUs to a single rendering task.  ...  This NVIDIA GRID system could be used as a rendering farm on a 10 gigabit network, eliminating the need for complex rendering architectures.  ... 
doi:10.11648/j.iotcc.20180602.11 fatcat:fw2kvioxi5fbnniq6yz7khi3ve

Toolkit for parallel image processing

Jeffery M. Squyres, Andrew Lumsdaine, Robert L. Stevenson, Hongchi Shi, Patrick C. Coffield
1998 Parallel and Distributed Methods for Image Processing II  
In this paper, we present the design and implementation of a parallel image processing software library (the Parallel Image Processing Toolkit).  ...  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.  ...  The PIPT thus incorporates a hierarchical model of parallelism that allows it to function on a single workstation, a single SMP, or a cluster of workstations and/or SMPs.  ... 
doi:10.1117/12.323475 fatcat:qqcx7xarunbtppsygvwjrfarvy

Performance Modeling and Evaluation of MPI

Khalid Al-Tawil, Csaba Andras Moritz
2001 Journal of Parallel and Distributed Computing  
In this paper, we use LogGP as a conceptual framework for evaluating the performance of MPI communications on three platforms: Cray-Research T3D, Convex Exemplar 1600SP, and a network of workstations (  ...  All of these implementations have a similar performance if compared on the same computing platform [18] .  ...  We like also to thank EPCC and the Department of Computer Science of the University of Edinburgh, United Kingdom, where part of this work was done.  ... 
doi:10.1006/jpdc.2000.1677 fatcat:m7ze7mle5jhstmvci5u5jepoq4

An efficient 3D grid based scheduling for heterogeneous systems

Anthony T. Chronopoulos, Daniel Grosu, Andrew M. Wissink, Manuel Benche, Jingyu Liu
2003 Journal of Parallel and Distributed Computing  
We implement our algorithms on a dedicated network of workstations (using MPI) and test them with a CFD simulation code (TURNS-Transonic Unsteady Rotor Navier Stokes). r  ...  The cost/performance ratio of networks of workstations has been constantly improving. This trend is expected to continue in the near future.  ...  We now consider the implementation on a network of heterogeneous workstations.  ... 
doi:10.1016/s0743-7315(03)00112-6 fatcat:43cn4jgllje7rfxvhbi2tfns5i

Lag in Multiprocessor Virtual Reality

Matthias M. Wloka
1995 Presence - Teleoperators and Virtual Environments  
While multiple, networked processors have been used to increase through-put, we concentrate on using multiple processors to reduce lag.  ...  To that end, we present a complete list of all possible lag sources in VR applications, review available lag reduction techniques, and investigate how these reduction techniques interrelate.  ...  Acknowledgements We thank Nate Huang and Tom Meyer for implementing parts of the example application, Bob Zeleznik for helpful discussions, our Sun sponsor Barry Medo for speedy resolution of hardware-speci  ... 
doi:10.1162/pres.1995.4.1.50 fatcat:rwyzft2xxretdg7p2626qug2li

Quantifying the Performance Differences between PVM and TreadMarks

Honghui Lu, Sandhya Dwarkadas, Alan L. Cox, Willy Zwaenepoel
1997 Journal of Parallel and Distributed Computing  
We compare two systems for parallel programming on networks of workstations: Parallel Virtual Machine (PVM), a message passing system, and TreadMarks, a software distributed shared memory (DSM) system.  ...  The rst is an 155Mbps ATM network with eight Sparc-20 model 61 workstations the second is an eight processor IBM SP/2.  ...  Introduction Parallel computing on networks of workstations has gained signi cant a t t e n tion in recent y ears.  ... 
doi:10.1006/jpdc.1997.1332 fatcat:3iuszu335va5fldran7ttdb46q
« Previous Showing results 1 — 15 out of 6,975 results