Abstracts - P2P and Grid Session
Title: P2P - Simulation and Reality
Sam Joseph, NeuroGrid Project, University of Tokyo
A large number of peer-to-peer (P2P) systems have been developed in recent years. This work is often motivated by the desire
to support increased scalability and robustness in the face of large numbers of users and highly dynamic network
environments. Just how robust and scalable these systems are is a hotly debated question. Information from network monitoring operations, simulations and analytical approaches are all used to support different sides of
the debate. Existing simulation systems have been joined by a large number of P2P specific simulation systems in
attempts to try and predict just how novel P2P approaches will perform under demanding real world
conditions. At the same time one must carefully ask the question of just how relevant these simulations are to the
reality of P2P systems operating in the real world. In this talk we will review a number of p2p simulation approaches,
with a focus on the extendible NeuroGrid P2P simulator. We will also look at how previously existing approaches to
network scalability in fields such a Grid-Computing overlap with th P2P simulation field.
Title: Logistical Networking: Provisioning Shared Resources
Both Peer-to-Peer and Grid systems seek to allow the sharing of resources
within a distributed community, but they have both strengths and limitations
reflecting their different orgins. The historical roots of Grid systems are
in the models of access and utilization controls that have been prevalent
in centralized, institutionally provisioned Data Centers, with emphasis on
security, accountability and billing. The historical roots of Peer-to-Peer
systems are in the models that have become prevalent in decentralized
networks, with emphasis on open, anonymous systems that emphasize openness,
anonymity and mutuality. The former has the advantage of supporting
individual resources on a large scale, while the latter excels in
scalability, resulting in an large and ubiquitous aggregate capacity.
Logistical Networking represents a middle path that combines many of the
strength of both of these well-known approaches. It is a technology for
provisioning shared resources, enabling both the efficiency of concentrated
institutional support for information technology infrastructure and the
autonomous allocation of those resources by applications and end users.
Title: On Death, Taxes and the Convergence of Peer-to-Peer and Grid Computing
Adriana Iamnitchi ,
Two supposedly new approaches to distributed computing have emerged in the
past few years, both claiming to address the problem of organizing
large-scale computational societies: peer-to-peer (P2P) and Grid computing.
Both approaches have seen rapid evolution, widespread deployment, successful
application, considerable hype, and a certain amount of (sometimes
warranted) criticism. The two technologies appear to have the same final
objective-the pooling and coordinated use of large sets of distributed
resources-but are based in different communities and, at least in their
current designs, focus on different requirements.
This talk takes some first steps towards comparing and contrasting P2P and
Grid computing. Basing the discussion whenever possible on the
characteristics of deployed systems, rather than the unverified claims
abundant in the literature, we review their target communities, resources,
scale, applications, and technologies. On the basis of this review, we draw
some initial conclusions concerning their interrelationship and future
evolution. In brief, we argue that (1) both are concerned with the same
general problem, namely the organization of resource sharing within virtual
communities; (2) both take the same general approach to solving this
problem, namely the creation of overlay structures that co-exist with, but
need not correspond in structure to, underlying organizational structures;
(3) each approach has made genuine technical advances but also has-in
current instantiations-crucial limitations, which we characterize
(simplistically, but still, we believe, usefully) as "Grid computing
addresses infrastructure but not yet failure; P2P addresses failure but not
yet infrastructure"; and (4) the complementary nature of the strengths and
weaknesses of the two approaches suggests that the interests of the two
communities are likely to grow closer over time.
Title: BOINC: A Platform for Public Distributed Computing
David P. Anderson, Space Sciences Lab, University of California at Berkeley
The majority of the world's processing power and storage
are in PCs owned by individuals.
Projects like GIMPs and SETI@home have demonstrated the
feasibility and potential of distributed computing using this resource.
We are currently developing BOINC,
a software platform to support projects of this sort.
BOINC will support a wide range of applications,
including those with extreme storage or communication requirements,
and will allow independent projects to share overlapping resource pools.
Title: Introduction to Grid Activities in the Asia
Yoshio Tanaka (AIST) and Peter Arzberger (UCSD, PRAGMA)
As in the United States and Europe, grid computing is regarded as a
viable next-generation computing infrastructure in Asia Pacific. Some
international efforts such as ApGrid and PRAGMA are trying to establish
sustained collaborations, build and deploy grid infrastructure, develop
grid middleware, and advance the use of grid technologies in
applications among a community of investigators working with leading
institutions in Asia Pacific. This talk introduces grid activities in
the Asia Pacific region, summarizes the current status of grid
technologies, and highlight applications in this region. We offer a
perspective on the convergence of grid and peer-to-peer approaches from
an application perspective.
Title: Jxta and Grid
Simon See (SUN, NTU)
Abstract- Recently there have been a lot of development of
applications using P2P and Grid technology. Each of them address a
certain set of issues. In this talk, we briefly discuss the how
Jxta is being used in some of the Grid Projects and what is needed to
for the first part, where a few p2p projects being undertaken across
AP countries would be introduced, Korea@Home (KISTI) project
and P3(Personal Power Plant) project lead by Kazuyuki Shudo (AIST, Japan, upon
the recommendation given by Dr. Satoshi Sekiguchi last month)
are being contacted and expected to be confirmed soon, in a few days.