L Smarr… - 1992 - Wiley Online Library From the standpoint of the average user, today's computer networks are extremely prim- itive
compared to other networks. While the national power, transportation, and telecom- munications
networks have evolved to their present state of sophistication and ease of use, computer ... Cited by 321 - Related articles - Leiden SFX Services - All 3 versions
[PDF] from psu.edu Leiden Fulltext SFXT Kielmann, RFH Hofman, HE Bal, A Plaat… - ACM Sigplan …, 1999 - portal.acm.org Writing parallel applications for computational grids is a challeng- ing task. To achieve good
performance, algorithms designed for lo- cal area networks must be adapted to the differences
in link speeds. An important class of algorithms are collective operations, such as ... Cited by 319 - Related articles - BL Direct - All 21 versions
[PDF] from psu.eduR Buyya, D Abramson… - 2001 - computer.org Page 1. A Case for Economy Grid Architecture for Service Oriented Grid Computing Rajkumar
Buyya, David Abramson, and Jonathan Giddy School of Computer Science and Software
Engineering Monash University Caulfield Campus, Melbourne, Australia ... Cited by 290 - Related articles - Leiden SFX Services - All 34 versions
[PDF] from psu.edu Leiden Fulltext SFXM Van Steen, P Homburg… - Concurrency, IEEE, 1999 - ieeexplore.ieee.org Scalable middleware requirements Our solution lies in the development of a wide-area distributed
system called Globe (visit http://www.cs.vu.nl/~steen/globe/). We aim to meet three major design
objectives: provide a uniform model for distributed computing, support a flexible implemen- Cited by 264 - Related articles - BL Direct - All 81 versions
[PDF] from psu.edu Leiden Fulltext SFXRAF Bhoedjang, T Ruhl… - Computer, 1998 - ieeexplore.ieee.org Modern high-speed local area networks offer great potential for communication- intensive
applications, but their perfor- mance is limited by the use of traditional communication
protocols, such as TCP/IP. In most cases, these protocols require that all network access ... Cited by 185 - Related articles - Library Search - BL Direct - All 24 versions
Leiden Fulltext SFXJ Maassen, R van Nieuwpoort, R Veldema… - ACM Sigplan …, 1999 - portal.acm.org Java offers interesting opportunities for parallel computing. In par- ticular, Java Remote Method
Invocation provides an unusually flex- ible kind of Remote Pmcedtue Call. Unlike RPC, RMI supports
polymorphism, which mquires the system to be able to download remote classes into a ... Cited by 159 - Related articles - BL Direct - All 11 versions
[PDF] from psu.edu Leiden Fulltext SFXG Pierre, M Van Steen… - IEEE Transactions on …, 2002 - computer.org AbstractTo improve the scalability of the Web, it is common practice to apply caching and replication
techniques. Numerous strategies for placing and maintaining multiple copies of Web documents
at several sites have been proposed. These approaches essentially apply a global ... Cited by 105 - Related articles - BL Direct - All 31 versions
[PDF] from psu.edu Leiden Fulltext SFXCGM Snoek, M Worring, JM Geusebroek… - IEEE Transactions on …, 2006 - computer.org Abstract—This paper presents the semantic pathfinder architecture for generic indexing of multimedia
archives. The semantic pathfinder extracts semantic concepts from video by exploring different
paths through three consecutive analysis steps, which we derive from the observation that ... Cited by 101 - Related articles - BL Direct - All 19 versions
[PDF] from ttu.edu.tw Leiden Fulltext SFXJ Waldo - Concurrency, IEEE, 1998 - ieeexplore.ieee.org REMOTE PROCEDURE CALL SYSTEMS have been around since Andrew Birrell and Greg
Nelson first proposed them in 1984.1 During the intervening 15 years, numerous evolution- ary
improvements have occurred in the basic RPC system, leading to improved ... Cited by 92 - Related articles - BL Direct - All 19 versions
[PDF] from psu.edu Leiden Fulltext SFXA Iosup, H Li, M Jan, S Anoep… - Future Generation …, 2008 - Elsevier While large grids are currently supporting the work of thousands of scientists, very little is known
about their actual use. Due to strict organizational permissions, there are few or no traces of grid
workloads available to the grid researcher and practitioner. To address this problem, in ... Cited by 80 - Related articles - All 13 versions
[PDF] from psu.edu Leiden Fulltext SFXFJ Seinstra, D Koelma… - Parallel Computing, 2002 - Elsevier The first part of the paper discusses implementation aspects of the parallel library, and shows
how sequential as well as parallel operations are implemented on the basis of so-called parallelizable
patterns. A library built in this manner is easily maintainable, as extensive code ... Cited by 79 - Related articles - BL Direct - All 25 versions
[PDF] from psu.eduHE Bal, A Plaat, MG Bakker, P Dozy… - … , 1998. IPPS/SPDP …, 1998 - ieeexplore.ieee.org Recent developments in networking technology cause a growing interest in connecting
local-area clusters of work- stations over wide-area links, creating multilevel clusters, or meta
computers. Often, latency and bandwidth of local-area and wide-area networks differ by ... Cited by 68 - Related articles - Leiden SFX Services - All 21 versions
[PDF] from cparity.com Leiden Fulltext SFXS Ben Hassen, HE Bal… - ACM Transactions on …, 1998 - portal.acm.org Many programming languages support either task parallelism or data parallelism, but few
lan- guages provide a uniform framework for writing applications that need both types of
parallelism. We present a programming language and system that integrates task and ... Cited by 63 - Related articles - BL Direct - All 9 versions
[PDF] from psu.eduAID Bucur… - 2003 - computer.org In systems consisting of multiple clusters of processors which are interconnected by relatively
slow communication links and which employ space sharing for scheduling jobs, such as our
Distributed ASCI1 Supercomputer (DAS), co- allocation, ie, the simultaneous allocation of ... Cited by 62 - Related articles - Leiden SFX Services - All 13 versions
Leiden Fulltext SFXAJC van Gemund - IEEE Transactions on Parallel and Distributed …, 2003 - computer.org AbstractPerformance prediction is an important engineering tool that provides valuable feedback
on design choices in program synthesis and machine architecture development. We present
an analytic performance modeling approach aimed to minimize prediction cost, while ... Cited by 59 - Related articles - BL Direct - All 13 versions
R Van Nieuwpoort, J Maassen, HE Bal… - Proceedings of the …, 1999 - portal.acm.org Java's support for parallel and distributed processing makes the lan- guage attractive for metacomputing
applications, such as parallel applications that run on geographically distributed (wide-area)
sys- tems. To obtain actual experience with a Java-centric approach to metacomputing, ... Cited by 59 - Related articles - Leiden SFX Services - All 6 versions
[PDF] from psu.eduCGM Snoek, M Worring, JM Geusebroek… - Proc. of TRECVID, 2004 - Citeseer Abstract This year the UvA-MediaMill team participated in the Feature Extraction and Search
Task. We developed a generic approach for semantic concept classification using the semantic
value chain. The semantic value chain extracts concepts from video documents based on ... Cited by 55 - Related articles - View as HTML - All 11 versions
Leiden Fulltext SFXKA Iskra, F Van der Linden, ZW Hendrikse… - ACM SIGOPS …, 2000 - portal.acm.org Abstract: Parallel programming on clusters of workstations is increasingly attractive, but dynamic
load balancing is needed to make efficient use of the available resources. Dynamite provides
dynamic load balancing for PVM applications running under Linux and Solaris. It supports ... Cited by 54 - Related articles - All 6 versions
[PDF] from psu.edu Leiden Fulltext SFXJ Liu, X Jin… - IEEE Transactions on Parallel and Distributed …, 2005 - computer.org Abstract—In this paper, we present a macroscopic characterization of agent-based load balancing
in homogeneous minigrid environments. The agent-based load balancing is regarded as agent
distribution from a macroscopic point of view. We study two quantities on minigrids: the ... Cited by 50 - Related articles - Library Search - BL Direct - All 14 versions
[PDF] from psu.eduA Plaat, HE Bal… - High-Performance Computer … - ieeexplore.ieee.org This paper studies application performance on systems with strongly non-uniform remote memory
access. In cur- rent generation NUMAs the speed difference between the slowest and fastest
link in an interconnect—the “NUMA gap”—is typically less than an order of magnitude, ... Cited by 48 - Related articles - Leiden SFX Services - All 18 versions
[PDF] from psu.eduAM Kermarrec, I Kuz, M Van Steen… - icdcs, 1998 - computer.org Despite the extensive use of caching techniques, the Web is overloaded. While the caching techniques
currently used help some, it would be better to use different caching and replication strategies
for different Web pages, depending on their characteristics. We propose a framework in ... Cited by 48 - Related articles - Leiden SFX Services - BL Direct - All 23 versions
[PDF] from psu.eduA Ganguly, A Agrawal, PO Boykin… - 2006 - computer.org Abstract—This paper describes WOW, a distributed system that combines virtual machine, overlay
networking and peer- to-peer techniques to create scalable wide-area networks of virtual workstations
for high-throughput computing. The system is architected to: facilitate the addition of ... Cited by 44 - Related articles - Leiden SFX Services - All 12 versions
[PDF] from psu.eduA Iosup… - 2006 - computer.org Grid computing is becoming the natural way to aggre- gate and share large sets of heterogeneous
resources. With the infrastructure becoming ready for the challenge, cur- rent grid development
and acceptance hinge on proving that grids reliably support real applications, and on cre- ... Cited by 43 - Related articles - Leiden SFX Services - All 14 versions
[PDF] from psu.edu Leiden Fulltext SFXFJ Seinstra… - Concurrency and Computation: …, 2004 - Wiley Online Library SUMMARY Although many image processing applications are ideally suited for parallel
implementation, most researchers in imaging do not benefit from high-performance computing
on a daily basis. Essentially, this is due to the fact that no parallelization tools exist that ... Cited by 39 - Related articles - BL Direct - All 21 versions
[PDF] from psu.eduF Breg… - Proceedings of the 2001 joint ACM- …, 2001 - portal.acm.org ABSTRACT Distributed computing has become increasingly popular in the high performance
community. Java's Remote Method Invocation (RMI) provides a simple, yet powerful method
for implementing paxallel algorithms. The performance of RMI has been less than ... Cited by 39 - Related articles - Leiden SFX Services - BL Direct - All 8 versions
Leiden Fulltext SFXAID Bucur… - IEEE Transactions on Parallel and …, 2007 - computer.org Abstract—Building multicluster systems out of multiple, geographically distributed clusters interconnected
by high-speed wide-area networks can provide access to a larger computational power and
to a wider range of resources. Jobs running on multiclusters and, more generally, in grids, ... Cited by 38 - Related articles - BL Direct - All 5 versions
[PDF] from psu.edu Leiden Fulltext SFXFJ Seinstra, D Koelma… - Parallel and Distributed …, 2004 - ieeexplore.ieee.org Abstract—A popular approach to providing nonexperts in parallel computing with an
easy-to-use programming model is to design a software library consisting of a set of preparallelized
routines, and hide the intricacies of parallelization behind the library's API. However, for ... Cited by 38 - Related articles - BL Direct - All 22 versions
[PDF] from psu.eduAID Bucur… - 2003 - computer.org In systems consisting of multiple clusters of proces- sors which employ space sharing for scheduling
jobs, such as our Distributed ASCI1 Supercomputer (DAS), co- allocation, ie, the simultaneous
allocation of processors to single jobs in multiple clusters, may be required. In this pa- per ... Cited by 36 - Related articles - Leiden SFX Services - All 13 versions
[PDF] from psu.eduS Banen, A Bucur… - Job Scheduling Strategies for Parallel …, 2003 - Springer Abstract. In systems consisting of multiple clusters of processors in- terconnected by relatively
slow network connections such as our Dis- tributed ASCI Supercomputer (DAS), applications
may benefit from the availability of processors in multiple clusters. However, the ... Cited by 35 - Related articles - Leiden SFX Services - BL Direct - All 10 versions
[PDF] from psu.eduAID Bucur… - Job Scheduling Strategies for Parallel …, 2001 - Springer Abstract. In systems consisting of multiple clusters of processors interconnected by relatively
slow connections such as our Distributed ASCI1 Supercomputer (DAS), jobs may request
co-allocation, ie, the simultaneous allocation of pro- cessors in different clusters. The ... Cited by 35 - Related articles - Leiden SFX Services - BL Direct - All 16 versions
[PDF] from psu.edu Leiden Fulltext SFXASZ Belloum, DL Groep, ZW Hendrikse… - Future Generation …, 2003 - Elsevier The Grid-based Virtual Laboratory AMsterdam (VLAM-G) provides a science portal for distributed
analysis in applied scientific research. By facilitating access to distributed compute and information
resources held by multiple organizations, and providing remote experiment control, data ... Cited by 33 - Related articles - All 14 versions
[PDF] from psu.eduAID Bucur… - 2003 - computer.org In systems consisting of multiple clusters of proces- sors which employ space sharing for scheduling
jobs, such as our Distributed ASCI1 Supercomputer (DAS), co- allocation, ie, the simultaneous
allocation of processors to single jobs in multiple clusters, may be required. In studies of ... Cited by 31 - Related articles - Leiden SFX Services - All 12 versions
[PDF] from psu.eduJW Romein, A Plaat, HE Bal… - Proceedings of the National …, 1999 - aaai.org This paper introduces a new scheduling algorithm for parallel single-agent search, transposition
table driven work schedul- ing, that places the transposition table at the heart of the par- allel
work scheduling. The scheme results in less synchro- nization overhead, less processor ... Cited by 29 - Related articles - BL Direct - All 30 versions
[PDF] from psu.edu Leiden Fulltext SFXRAF Bhoedjang, K Verstoep, T Rühl… - ACM SIGARCH …, 2000 - portal.acm.org ABSTRACT We systematically evaluate the performance of five implementa- tions of a
single, user-level communication interface. Each im- plementation makes different architectural
assumptions about the reliability of the network hardware and the capabilities of the net- ... Cited by 29 - Related articles - BL Direct - All 21 versions
[PDF] from psu.eduG van Albada, J Clinckemaillie, A Emmen… - High-Performance …, 1999 - Springer Workstations have become ubiquitous in many organisations. By their nature, they are often
used intensively during normal working hours, and are often largely idle other- wise. They represent
a huge reservoir of computing capacity that can be used much more efficiently. Thus, we ... Cited by 27 - Related articles - Leiden SFX Services - BL Direct - All 16 versions
[PDF] from psu.eduG Allen, T Dramlitsch, T Goodale… - … Computing and the …, 2001 - ieeexplore.ieee.org Page 1. Early Experiences with the EGrid Testbed Gabrielle Allen, Thomas Dramlitsch,
Tom Goodale, Gerd Lanfermann, Thomas Radke, Ed Seidel Max-Planck-Institut fur
Gravitationsphysik, Potsdam, Germany Thilo Kielmann ... Cited by 26 - Related articles - Leiden SFX Services - All 31 versions
[PDF] from psu.edu Leiden Fulltext SFXKA Iskra, RG Belleman… - Concurrency and …, 2002 - Wiley Online Library SUMMARY The paper provides an overview of an experimental, Grid-like computing
environment, Polder, and its components. Polder offers high-performance computing and interactive
simulation facilities to computational science. It was successfully implemented on a wide- ... Cited by 26 - Related articles - BL Direct - All 7 versions
[PDF] from psu.edu Leiden Fulltext SFXFJ Seinstra… - IEEE Transactions on Parallel and …, 2002 - computer.org AbstractOne of the most fundamental problems automatic parallelization tools are confronted
with is to find an optimal domain decomposition for a given application. For regular domain problems
(such as simple matrix manipulations), this task may seem trivial. However, ... Cited by 25 - Related articles - BL Direct - All 22 versions
