CCGrid 2001: Brief Summary + CCGrid 2002: call for participation

Rajkumar Buyya rajkumar at
Sun Jun 3 19:20:21 PDT 2001

Dear All,

Please find enclosed a brief summary on our CCGrid 2001 symposium written by
Craig Lee from Aerospace Corporation. 

The presentation slides of all keynote, tutorials, invited, and industry talks are
available for download from the CCGrid 2001 web site:
(go to Final program and click on respective titles to download PPT version of slides).

CCGrid 2002 is scheduled to be held in in Berlin, Germany, May 21-24, 2002. 
Full details (call for papers, tutorials, workshops, posters etc.) are available at: |

Thanks for your contributions that made CCGrid 2001 successful and we look forward for
your similar support for the next conference.

A Brief Summary of CCGrid 2001: Cluster and Grid Computing Symposium

Craig A. Lee, Aerospace Corporation, Los Angeles, USA.

The First IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGrid) was
held on May 15-18, 2001, in Brisbane, Australia. The meeting has been sponsored by the
IEEE Task Force on Cluster Computing (TFCC) and hosted by the Queensland University of
Technology (QUT). CCGrid focuses on the combined areas of clusters and Grid computing,
which share many related technical issues and are both areas of intense interest and rapid
growth. Cluster computing has enabled low-cost entry into supercomputing performance by
using clusters based on commodity components, such as processors and network
infrastructure. Grid computing borrows its name from the analogy with the electrical power
grid. The electrical power grid made electricity widely available and easy to use. The
"information power Grid" endeavours to make the discovery and sharing of information and
resources widely available and easy to use. Clusters and Grids share many communication,
scheduling, monitoring, and application development issues, with Grids being the most
general case since they can be heterogeneous and open-ended.

Following a traditional structure, the symposium consisted of seven keynote addresses and
invited talks, three tutorials, seven workshops, forty-eight technical papers, a poster
session, an industry track, and a panel. The keynotes covered the spectrum of important
cluster and Grid computing issues: Ian Foster of Argonne National Lab spoke on Grid
architecture ("The Anatomy of the Grid: Enabling Scalable Virtual Organizations"), Andrzej
Goscinski of Deakin University spoke on cluster organization and management ("Making
Parallel Processing on Clusters Efficient, Transparent, and Easy for Programmers"),
Satoshi Matsuoka of Tokyo Institute of Technology spoke on programming ("Grid RPC meets
Data Grid: Network Enabled Services for Data Farming on the Grid"), Greg Pfister of IBM
spoke on a new communication technology ("The Promise of InfiniBand for Cluster
Computing"), and finally, Bruce Maggs of Akamai spoke on content delivery ("Global
Internet Content Delivery"). The two invited speakers Gul Agha of University of Illinois
and Jeff Bradshaw of Boeing have spoke on Actors and Agents technologies for Terraforming
the Cyberspace.    

The keynotes set the tone for the rest of the symposium. The main symposium technical
tracks covered component and agent approaches, Grid computing, scheduling and load
balancing, message passing and communication, I/O and databases, performance evaluation,
distributed shared memory, and tools for management, monitoring and debugging. The seven
workshops presented more recent "work-in-progress" in areas closely related to the
technical tracks: agent based cluster and Grid computing, object and component technology
for cluster computing, quality of service for global computing, scheduling and
load-balancing on clusters, global computing on personal devices, distributed shared
memory, and cluster computing education.

The symposium concluded with a panel: "The Grid: Moving It to Prime Time" that was
moderated by David Abramson. Panellists included Satoshi Matsuoka, Craig Lee, Gul Agha,
and Bruce Maggs. Besides discussing the myriad of technical issues surrounding the
development of effective Grid computing in general, the panel discussed the even more
problematic issues of moving grids from the scientific and engineering communities to be
part of the mainstream-computing infrastructure that is enveloping the world. 

Grid computing has emerged as the predominant approach for wide-area, high-performance
computing, but other approaches, such as Peer-to-Peer Computing and CORBA, are also
emerging, and these technologies are motivated more by the business-to-business and
business-to-consumer markets. However, these application domains are faced with the same
fundamental problems, e.g., resource discovery, scheduling, security, etc., but the
solution spaces and potential implementations could be quite different and determined by
the commercial marketplace. Hence, the future of cluster and Grid computing will be
heavily influenced by how they co-evolve with these other global computing paradigms.

CCGrid 2001 was highly successful by any standards, and especially for a new symposium. It
attracted world-renown computer scientists from 28 countries with a high-quality program.
CCGrid 2002 has already been announced to be in Berlin, Germany, May 21-24, 2002. Full
details are available at |

More information about the Beowulf mailing list