|
|
|
Session Chair:
Hyun Chul Kim (KAIST, Korea)
Presentation Titles and Speakers:
1.Introduction of Logistical Networking (download
presentation)
Micah Beck (University of Tennessee, US)
2.Replica Management for IBP (download
presentation)
Tang Ming (NTU, Singapore)
3.
Distributed Data Storage (download
presentation)
Ludek Matyska (Masaryk University, Czech Republic)
4.Web 100 Project on Logistical Networking (download
presentation)
Jim Ferguson
(University of Illinois, Urbana-Champaign, US)
Description: Logistical Networking can be defined
as the global optimization and scheduling
of data storage, data movement, and computation. It is a
technology for shared network storage that allows an easy
scaling in terms of the size of the user community, the
aggregate quantity of storage that can be allocated, and the
distribution breadth of service nodes across network borders. As
an advanced engineering testbed for advanced applications, we
expect that it will bring a set of
research/engineering/application opportunities to the
communities on a high performance research and education
network.
In this session, after introducing the base concepts and visions
of Logistical Networking, its current development and deployment
status (across APAN/Internet2 networks), as well as applications
on it will be described. We conclude this session by presenting
and discussing the research and engineering
issues/opportunities to address, and possible applications that
are expected to gain benefits from it (and/or to be
enabled/realized by it).
Presentation Abstracts:
1. Introduction of Logistical Networking
Abstract:
This talk will explore the concepts and
mechanisms underlying Logistical Networking, a revolutionary
architectural approach to communication that synthesizes
elements of storage and wide area networking systems that are
traditionally considered orthogonal. Logistical Networking is
modeled on IP networking, and so its architecture is a stack
with physical media and OS drivers at the bottom; an innovative
layer that enables the scalable sharing of storage called the
Internet Backplane Protocol (serving a function analogous to
IP); and then the exNode, a tool for aggregating resources and
enabling valuable end-to-end services such as reliability, high
performance, and security (serving a function analogous to
TCP). Researcher at the Logistical Computing and
Internetworking Laboratory at the University of Tennessee have
been pursuing the development of these mechanisms and their
integration into higher level middleware and application-level
tools in an effort to create a new architecture for scalable
computing in the wide area based on the successful architectural
approach of the Internet. This new architecture offers a new
framework for advanced Internet applications of all kinds, from
scientific Grid computing to collaborative work to multimedia
content delivery.
2. Replica Management for IBP
Abstract: One of the major tasks in Data Grid is the
distribution and storage of
Data. In this presentation we will report on the development of
a
prototype Data Grid with replica capability. It integrates the
Internet Backbone Protocol and with Globus Replica Catalog. The
Internet
Backplane Protocol (IBP) is a middleware for sharing storage
resources
in world wide and it is implemented as part of the Logistical
Networking. IBP provides a basic scalable infrastructure for the
Data
Grid. Currently IBP lacks the replica management service, so the
replica stored in IBP are not sharable to public. In this
project, we take IBP
as the underlying system of the Data Grid and put forward an
implementation of replica management service, so that the data
resources are shareable to public. The Globus Replica Catalog is
integrated to
the system to provide the replica location service. And more
sub-systems,
exNode File Warehous and IBP Replica Management API, are
developed.
With the Replica Management System for IBP, the storage and data
resources
can be easily shared among distributed communities in world
wide, and
IBP can also be applied to support Grid computing more
effectively.
3. Distributed Data Storage
Abstract: DiDaS is a national-wide
testbed for both research and applications of
the Internet backplane protocol and related approaches to
Logistic
networking. A set of 7 storage depots with a total capacity of
around
10TB is spread around the Czech Republic and connected directly
to the high
speed backbone. This infrastructure is used to store and provide
fast
acces to the large collection of video data (the streaming
storage),
a collection of digital maps, some web harvesting results and
will also
be used as a large "intelligent" cache for data used by the
national
Grid. The research includes issues on security and reliability
on top of
underlying best effort protocol, and also on scheduling data
transfers
(for intelligent caching) with respect to the network
conditions (actual
and to some extent also predicted).
4. Web 100 Project on Logistical Networking
Abstract: It is common for applications, hosts,
researchers, and others to be unable to take advantage of
national and international high performance network growth in
bandwidth and accessibility. Without expert attention from
network engineers, users are unlikely to achieve even 50 Mbps
single stream TCP transfers, despite the fact that the
underlying network infrastructure can support data rates of 622
Mbps or more. This Web100 project was created to produce a
complete host-software environment that will run common TCP
applications at 100% of the available bandwidth, regardless of
the magnitude of a network's capability. The impact of this
type of technology on logistical networking will be discussed.
|
