Home Up Feedback Contents Search

 
Research

 

 

 

 

 

Sponsor Links

IBM Virtualization Engine Grid Toolbox for Multiplatforms - ( v. 1.x ) - media
HP/Compaq Proliant BL20p G3 Blade Server
IBM eserver BladeCenter HS20 8832 - Xeon 3.2 GHz
logo_88x31
HP StorageWorks Modular SAN Array 1000, includes one controller and one single-port Fibre Channel I/O Module
Hotwire.com
Intel Server Compute Blade SBXL52 - no CPU
Red Hat Linux Advanced Server
Intel Cluster Math Kernel Library for Linux - ( v. 7.x ) - complete package
Dell Outlet
Intel Blade Server Chassis SBCE - desktop - 7 U
iBook G4
HP Installer Kit for Linux - media
HP StorageWorks Continuous Access EVA - ( v. 1 ) - complete package
HP Fabric Manager Enterprise - ( v. 4.x ) - complete package
IBM Cluster Systems Management Base for Xlinux/EServer - ( v. 1.4 ) - media
HP F500 Cluster for EVA Basic
S/W Integration Kit for HP OpenView NNM SNMP MGMT
HP StorageWorks 300mx MO Jukebox 2 Drives , 291.2 GB
Novanet Microsoft Clusters
HP StorageWorks Magneto-Optical Storage 2200mx , 2.17TB

 
 

This section highlights research papers about grids and grid computing. It includes publications from academic and research organizations.

 

A Unified Relational Approach to Grid Information Services
Peter A. Dinda (Northwestern University), Beth Plale (Indiana University)

The authors defines a Grid Information Service (GIS) as a database of entities within a grid, including organizations, computers, routers, software and instruments. The GIS includes objects, relationships between objects, and systems need to update them. The authors suggest hierarchical structures (such as LDAP) will give way to relational databases as grids become more complex and contain more objects.

Cactus Application: Performance Predictions in Grid Environments
Matei Ripeanu, Adriana Iamnitchi, Ian Foster (University of Chicago)

The authors evaluated the performance of two types of grids for tightly-coupled scientific applications that require huge amounts of memory processor time. One type was a grid of supercomputers and another was a grid of one million workstations. The authors found the latter architecture was about 30 times faster than the supercomputer grid.

Distributed Database Management Systems and the Data Grid
Heinz Stockinger (CERN, European Organization for Nuclear Research)

This paper presents a Data Grid capable of managing data persisted in object-oriented databases. The author describes massive data collections used for high-energy physics experiments such as BaBar, CMS and Atlas and suggests an approach for managing petabytes of data. He also describes replication scenarios and the current focus of distributed database research.

Fujitsu

From Open Grid Services Infrastructure to WS-Resource Framework: Refactoring and Evolution
Ian Foster, Steve Tuecke (Argonne National Laboratory/Globus Alliance), David Snelling (Fujitsu),
Don Ferguson, Jeff Frey, Steve Graham, Tom Maguire (IBM),  Karl Czajkowski (USC ISI/Globus Alliance)

The Open Grid Services Infrastructure (OGSI) specification defined extensions to XML Schema and WSDL for supporting stateful services. It was re-factored in 2004 to take advantage of web services standards such as WS-Addressing and WS-Notification.

GrADSolve - a Grid-based RPC system for Remote Invocation of Parallel Software
Satish S. Vadhiyar, Jack J. Dongarra (University of Tennessee)

This paper describes an RPC system that supports execution of parallel applications that are dynamically chosen based on load characteristics of machines in a grid. It provides execution traces and data staging.

Implementations of a Service-Oriented Architecture on top of Jini, JXTA and OGSI
Nathalie Furmento, Jeffrey Hau, William Lee, Steven Newhouse, John Darlington (Imperial College London)

This paper presents a Service-Oriented Architecture (SOA) that's the main basis for ICENI Grid
middleware. The authors compare implementations built on top of Jini, JXTA and the Open Grid Services Infrastructure (OGSI).

Mobile Interfaces to Computational, Data and Service Grid Systems
Mauro Migliardi (University of Genoa), Muthucumaru Maheswaran, Balasubramaniam Maniymaran, Paul Card, Farag Azzedin (University of Manitoba and TRLabs)

The authors examine mobile access to grids and issues such as resource heterogeneity, load balancing, quality of service, terminal mobility and user mobility. They describe
the HARNESS system and Invisible Network (InviNet) architecture.

The Anatomy of the Grid: Enabling Scalable Virtual Organizations
Ian Foster, Steven Tuecke (Argonne National Laboratory), Carl Kesselman (University of Southern California)

The authors explain grids as a solution for sharing resources and applications among institutions and individuals that compose virtual organizations. This is one of the most-cited papers about grid computing.

The Physiology of the Grid: An Open Grid Services Architecture for Distributed Systems Integration
Ian Foster, Steven Tuecke (Argonne National Laboratory), Carl Kesselman (University of Southern California), Jeffrey M. Nick (IBM)

Describes OGSA grid service semantics and techniques for creating, naming, and discovering grid service instances. It also provides multiple protocol bindings for service instances.

Toward a Framework for Preparing and Executing Adaptive Grid Programs
Ken Kennedyα, Mark Mazina, John Mellor-Crummey, Keith Cooper, Linda Torczon (Rice University).
Fran Berman, Andrew Chien, Holly Dail, Otto Sievert (University of California, San Diego), Dave Angulo, Ian Foster (University of Chicago), Dennis Gannon (Indiana University), Lennart Johnsson (University of Houston), Carl Kesselman (USC/Information Sciences Institute), Ruth Aydt, Daniel Reed (University of Illinois, Urbana-Champaign), Jack Dongarra, Sathish Vadhiyar (University of Tennessee), Rich Wolski (University of California, Santa Barbara)

Describes the program execution framework developed by the Grid Application Development Software (GrADS) Project. The goal is to optimize resource allocation for grid applications and support adaptive reallocation based on changes in performance.

Towards Virtual Networks for Virtual Machine Grid Computing
Ananth I. Sundararaj, Peter A. Dinda (Northwestern University)

The authors suggest that grid application software and middleware are complex. To simplify the configuration management issue, the authors suggest using simple middleware hosted by virtual machines. The authors descibe VNET, a virtual network tool they developed for local area and wide area networks. VNET is a virtual private network that implements a virtual LAN. It simplifies migration of machines from one site to another without having to re-configure IP addresses and routing information. VNET is available for VMware.

 

Home ] Up ]

Send mail to with questions or comments about this web site.
Copyright © 2005-7 GridSummit.com
Last modified: 10/30/07