About GGF · Overview · Who's
Involved? · People · Structure &
Process · History &
Background · Origin · Document
Process
GGF Intellectual
Property Policies
Get
Involved · How to Get
Involved · Join · Working
Groups · Membership ·
Sponsor
Membership · Sponsors ·
Scholarships ·
Grid
Initiatives
News
& Events · Announcements · What's
New? · In the Press · Related
Events · Newsletters
Contact the GGF ·
General
Inquiries · GGF People ·
GGF
Offices
NEXT GGF
Event GGF5 21-24 July 2002 Edinburgh, Scotland,
UK
|
|
|
Global
Grid Forum 5: BOF Program |
BOF: Appliance Aggregation
Date/Time |
Tue, Jul 23,
2002: 4:30pm - 6:30pm |
Location |
Carrick (50) |
Area |
Peer 2 Peer |
Description |
Dejan Milojicic, Hewlett
Packard |
Note |
Today, people are overloaded with IT appliances
battling for the real estate of their belt (phone, PDA,
camera, GPS) desk (displays, keyboards, mice), and closet
(surrogate servers, storage). All of these appliances are real
islands today, with separate power, hardware, interfaces,
identity, ownership, state, and so forth. By employing the
techniques of aggregation, we can establish a common
ownership, shared state, shared applications, and shared
services for these appliances. An appliance aggregation
architecture would make appliances work better together by
removing the barriers and better leveraging their resources.
It may even result in minimizing or eliminating some of the
appliances. Using one or a few appliances, a user can leverage
environmental appliances by aggregating them with his personal
ensemble of appliances. In this way users can limit the effort
in using and managing the appliances. Please join us for a
birds-of-feathers session at GGF to explore whether there is a
broader interest in appliance aggregationarchitecture among
the GGF members and their companies. We would also like to
consider starting a research group (RG) on this topic. The RG
is intended to start under the p2p technical area, but it
should encompass the GGF in
general. |
Name |
Andrew A.
Chein |
Email |
achien@entropia.com |
WebSite |
|
BOF: Grid Scheduling Architecture
Date/Time |
Tue, Jul 23,
2002: 4:30pm - 6:00pm |
Location |
Harris 1 (40)
|
Area |
Scheduling
& Resource Management |
Description |
This group
will define the architecture of a grid scheduler, and
interactions between a grid scheduler and other components
like the GSI, and local schedulers or NW management. |
Note |
Based on the
Scheduler Attribute Document, this working group has the
object to define the architecture of a grid scheduler. This
definition shall serve as basis for a first implementation of
such a grid scheduler. Specifically, we want to discuss the
interaction between a grid scheduler and other components like
the grid information system, local resource management systems
and network management systems. Additionally, we want to
address the relationship between hardware resource management
and data management. In this session we present a draft for a
working group charter and a rough first draft of a document
describing this architecture. |
Name |
Uwe
Schwiegelshohn |
Email |
mailto:Uwe.Schwiegelshohn@udo.edu |
WebSite |
http://www.mcs.anl.gov/~jms/ggf-sched/GGF5/ |
BOF: Scheduling Optimization
Date/Time |
Tue, Jul 23,
2002: 7:30pm - 8:30pm |
Location |
Carrick (50) |
Area |
Scheduling
& Resource Management |
Description |
We are
proposing a new research group in scheduling optimization. |
Note |
The aim of
this research group is the investigation of the available
scheduling techniques in the framework of the Grid
infrastructure. Possible goals for this group include case
studies of existing optimization techniques used in current
schedulers, agreement on terms to discuss optimization
techniques, comparison of optimization techniques used in
current systems, etc. |
Name |
Vincenzo Di
Martino |
Email |
mailto:vincenzo.dimartino@caspur.it |
WebSite |
www.mcs.anl.gov/~jms/ggf-sched/GGF5/
|
BOF: Grid Checkpoint Recovery Proposed
WG
Date/Time |
Tue, Jul 23, 2002: 6:00pm -
7:00pm |
Location |
Harris 1 (40) |
Area |
Applications &
Models |
Description |
The proposed GGF Working Group on Grid Checkpoint
Recovery (GridCPR) seeks to bring together a variety of Grid
users, developers and Grid resource providers to develop and
establish standard methods, APIs, and protocols for performing
Checkpointing and Job Recovery in Grid environments. In
addition to providing a basic form of fault tolerance, Grid
Checkpoint Recovery will be necessary to facilitate migration
of computations among available resources. A draft charter for
the working group is available at
http://gridcpr.psc.edu/GGF/GridCPR-WG-draft.txt |
Note |
The Grid Checkpoint Recovery Proposed WG BoF will
focus on - Defining the scope and activities of the
proposed Working Group - Revising/strengthening the charter
for the proposed Working Group - Planning subsequent
meetings and activities |
Name |
Derek Simmel |
Email |
dsimmel@psc.edu |
WebSite |
http://gridcpr.psc.edu/GGF/ |
BOF: Semantic Grid Proposed RG
Date/Time |
Tue, Jul 23,
2002: 7:30pm - 8:30pm |
Location |
Harris 1 (40) |
Area |
Architecture |
Description |
This group will look at issues concerned with
the use of Semantic Web tools by Grid Applications. |
Note |
Goals: Many
grid applications are set to benefit from semantic web tools
and techniques. The semantic web includes standards and tools
for immediate use (e.g RDF), ongoing activities (such as the
W3C Web Ontology Working Group) and an active community of
researchers. This RG provides a forum to track semantic web
community activities, determine relevance to grid activities,
provide a route for transfer of information and ideas between
the communities and coordinate activities as
appropriate.
Projected Tasks: 1) Track semantic web
activities 2) Coordinate and interact with other GGF groups
(for example, Data Access and Integration Services) 3) Feed
requirements and experiences from the grid community back to
the semantic web community 4) Operate a community web
portal to facilitate the bridge 5) Create focused Working
Groups as appropriate (e.g. one can envisage a working group
relating to ontologies).
Lifetime: This RG has an
indefinite life time. |
Name |
Dave De Roure |
Email |
dder@ecs.soton.ac.uk |
WebSite |
None |
BOF: Grid Information
Retrieval WG
Date/Time |
Tue, Jul 23,
2002: 6:00pm - 7:00pm |
Location |
Harris 2 (40) |
Area |
Performance
and Information Services |
Description |
This WG would look at requirements and a
reference implementation for information retrieval on the Grid
(documents etc.) The session will be led by Kevin
Gamiel. |
Note |
Mission:
The GridIR WG/RG will focus on
establishing the requirements, specifications, reference
implementations and best practices in supporting Information
Retrieval (IR) services on the Grid. Grid IR services will be
needed by users, applications and portals to provide
documents, document extracts, answers or other data items to
satisfy information needs.
Goals:
The GridIR
WG/RG will focus on the following:
1. Establish the
requirements for Grid IR services:
GridIR will be
defined as a set of grid services which, together, can be used
as an IR system, including:
Harvesters, to gather
network-based documents Indexers, to build data- and
file-structures for retrieval Index processors, to
determine post-indexing term and document weights Query
processors, to take user queries and gather results
Integrators, for ranking results from different sources
Renderers, to take results and organize or present them
Many other sub-systems and control systems GridIR will
also need to impose requirements on the IR service specific to
the Grid, including: Rapid update schedules for datasets
Federation of datasets from multiple sources Enabling
local policy for dataset content access, based on Grid
security infrastructure Sophisticated localized indexing
and query processing appropriate for each dataset
Sophisticated post-hoc results ranking Efficient use
of computational resources (e.g., multiple harvesters feeding
one indexer) Multimedia capabilities (incorporation of
special-purpose IR systems into one meta-system) Rapid
rendering and context-switching, including data visualization
of results and multiple 'views' of data b ased on different
user profiles Consensus-based results generation from
multiple retrieval algorithms to select best-of-breed
algorithms
2. Define a set of GridIR
specifications:
The Open Grid Services Architecture
(OGSA) along with technlogies such as the Web Services Flow
Language (WSFL) provide a framework for linking loosely
coupled grid services together to form more advanced services.
Though these technologies provide the infrastructure, each
service description must be created by stakeholder communities
to ensure required functionality. The GridIR WG/RG will
develop an overarching IR architecture, will detail
service-level requirements, will establish independent service
models, and develop interface specifications for the various
independent IR-related services, all with an eye towards tying
those services together into an integrated whole. The WG/RG
will work to develop a plug-and-play type architecture for
GridIR where the Grid infrastructure enables rapid integration
of standards-compliant IR modules. In many cases, GridPipes
will allow communication between modules (e.g., for multiple
harvesters feeding an indexer).Anticipated individual services
include crawlers, indexers, search and presentation engines.
3. Support and Evaluate GridIR Reference
Implementations
There are numerous investigation areas
for the reference implementation for GridIR specifications.
The reference implementations will address many of the
following IR considerations: Extremely large collections
(billions of documents) Documents in plain text, HTML, XML
Multimedia documents (video, audio, other non-text
formats) Documents in multiple languages; queries in
multiple languages All variety of harvesting methods
Numerous fundamental IR algorithm components (Boolean;
Vector Space Model, probabilistic IR, Page Rank, Latent
Semantic Indexing...) Flexible local policy for what
documents are allowed Sub-document retrieval, linguistic
approaches, question answering Long and short queries;
document filtering Solutions for most of the IR techniques
are available, although some do not scale well or are less
amenable to the distributed processing of the Grid.
GridIR will benefit from past experiences in networked
IR. For example, Z39.50 offers the ability to send a query to
multiple IR engines. GridIR will take Z39.50 further by
layering IR on the Grid security and authentication
infrastructure, and by providing sophisticated techniques for
merging and ranking the results from the engines.
To
support the evaluation of the reference implementations, the
GridIR WG/RG will promote the development of test suites that
can be used to validate an implementation and provide the
basis for comparing them.
4. Establish Best Practices
for GridIR
The GridIR WG/RG will establish best
practices for GridIR implementations and use by collecting and
disseminating experiences. Furthermore, the GridIR WG/RG will
ensure that the best practices conform with the other Grid
Services groups which define services that will be needed to
implement GridIR as well as the Portals, Users, and
Applications groups that will use the GridIR services. |
Name |
Kevin Gamiel |
Email |
kgamiel@cnidr.org |
WebSite |
www.gridir.org/charter.html |
BOF: Service Management
Frameworks RG
Date/Time |
Tue, Jul 23, 2002: 7:30pm -
8:30pm |
Location |
Harris 2 (40) |
Area |
Architecture |
Description |
Topics: Grid/Web Services
Support Frameworks;Service Discovery AP; JXTA + Jini; A
test-bed for deployment and testing of SMFs;
co-ordination of immediate joint efforts with OGSI, P-2-P,
etc. new name: Service Management
|
Note |
|
Name |
Steven Newhouse |
Email |
sjn5@doc.ic.ac.uk |
WebSite |
None |
BOF: Data Transport Proposed WG
Date/Time |
Tue, Jul 23,
2002: 6:00pm - 7:00pm |
Location |
Ochil 1 (50) |
Area |
Data |
Description |
This group
will look at issues in secure, robust, high performance
transport of data. |
Note |
Goals: Provide a forum where parties
interested in the secure, robust, high speed transport of data
in the wide area and related technologies can discuss and
coordinate issues. Note that this RG and GGF in general do not
"promote" research, they simply provide a forum for
coordination and standardization. For this reason, we make no
exclusion as to what can be discussed. Ultimately, the test
is whether or not sufficient people show interest in a topic.
This RG should be a WG "factory" spawning off WGs in areas
where sufficient common interest is shown.
This RG is
NOT relegated to GridFTP only. It will clearly need to
coordinate with the High Performance Networking and
Replication Groups. There is also a high probability that
some work developed here MAY need to go to IETF for approval.
GridFTP is an example. There is an FTP working group within
IETF and once we decide we have the protocol right AND we
decide it is appropriate for general internet usage (this is
not a forgone conclusion) then we will advance it though the
IETF. Similar decisions will be needed for other work spawned
from this group.
Projected Tasks: 1) Develop
appropriate scenarios to use as guidelines and test
cases. 2) Coordinate and interact with replication and high
performance networking RGs 3) Agree on operations necessary
at remote sites. This will help us understand interfaces and
possible protocol support needed. 4) Requirements for a
General Transport Protocol (DOC, possible WG) 5) Protocol
improvements to V1.0 of GridFTP based on #1 (DOC, Possible
WG) 6) Implementation requirements for a Transport system
(possible DOC, possible WG) 7) Performance Issues related
to high speed, bulk data transport (presentations) 8)
Evaluation of non-TCP based bulk data transport
(presentations)
Lifetime: This RG has an indefinite
life time. While it holds value as a forum for discussions and
presentations, it will also strive to produce focused WGs that
have a limited lifetime and produce standards track documents.
|
Name |
Bill Allcock |
Email |
mailto:allcock@mcs.anl.gov |
WebSite |
http://www.gridforum.org/meetings/ggf5/bofs.htm |
BOF: Common Schemas for Interoperability - has
been combined with Common Information Model (CIM)
WG
Date/Time |
Tue, Jul 23,
2002: 4:30pm - 6:00pm |
Location |
Tinto (200)
[note new location, combined with
CIM) |
Area |
Performance
and Information Services |
Description |
We would like
to discuss the possibility of starting a working group to
extend the GLUE-schema work for use in the broader GGF
community. |
Note |
|
Name |
Jenny Schopf,
Brian Tierney |
Email |
mailto:jms@mcs.anl.gov |
WebSite |
http://www.hicb.org/glue/glue-schema/schema.htm |
BOF: Grid Economic Brokering Architecture
WG
Date/Time |
Tue, Jul 23,
2002: 4:30pm - 6:00pm |
Location |
Ochil 1 (50) |
Area |
Scheduling
& Resource Management |
Description |
Steven
Newhouse, LeSC Technical Director and Jon MacLaren, Manchester
Computing |
Note |
Interfaces needed to extensibly support a
variety of negotiating mechanisms for the charging
of Grid Services deployed within OGSA.
|
Name |
Steven
Newhouse |
Email |
mailto:sjn5@doc.ic.ac.uk |
WebSite |
www.mcs.anl.gov/~jms/ggf-sched/GGF5 |
BOF: OGSA Resource Usage Service Proposed
WG
Date/Time |
Tue, Jul 23,
2002: 7:30pm - 8:30pm |
Location |
Tinto (200) |
Area |
Scheduling
& Resource Management |
Description |
Steven
Newhouse |
Note |
Purpose:
To define an OGSA compatible Resource Usage Service and
associated client side and management tools for deployment
within an OGSA hosting environment. This service will track
resource usage (accounting in the traditional UNIX sense) and
will not concern itself with payment for the use of the
resource. Goal: To enable the tracking of resource
usage within Grid Services deployed within an OGSA
environment. As the 'resources' that need to be tracked (e.g.
CPU, time, memory) may be vary between services and over time
an extensible schema will be used to structure this
information. Activity: a) Definition of service
interface to add & search records for accounting
information. b) Defining a security model that protects
and individuals right to privacy relating to how they used a
service. c) Define a (minimal but extendsible) set of
commonly understood attributes to describe resource usage.
(NB: Potentially leverage existing work in the Accounting RG
and the Scheduling Dictionary WG.) Plan: · GGF 5:
Initiate discussion of this activity with a BOF and a move to
produce a WG charter. · GGF 6: Contribute resource usage
cases (from UK Computational Economy project) to build on
exisiting information in this area within GGF. · GGF 6:
Provide for detailed discussion an intitial specification of
this service. (NB: An alpha implementation may be available
from UK e-Science activity.) · GGF 7: Present a revised
specification for further discussion. (NB: A beta
implementation will be available from UK e-science activity.)
· GGF 8: Complete initial discussion of service
specification for movement through GGF documents process. (NB:
A reference implementation of this Resource Usage Service
should be available for deployment within an OGSA
environment.) |
Name |
Steven
Newhouse |
Email |
sjn5@doc.ic.ac.uk |
WebSite |
www.mcs.anl.gov/~jms/ggf-sched/GGF5 |
BOF: OGSA
Roadmap
Date/Time |
Tue, Jul 23,
2002: 6:00pm - 7:00pm |
Location |
Pentland
Auditorium (600) |
Area |
Architecture |
Description |
A Open Grid
Service Architecture BOF, aimed at introducing and defining a
new OGSA WG, that would look at things beyond GS Spec |
Note |
|
Name |
Steve Tuecke |
Email |
mailto:tuecke@mcs.anl.gov |
WebSite |
None |
BOF: Usage Record
WG
Date/Time |
Tue, Jul 23, 2002: 6:00pm - 7:00pm |
Location |
Carrick (50) |
Area |
Scheduling and Resource Management |
Description |
In order for resources to be shared, sites must
be able to exchange basic usage data in a common format. This
group will define a common usage record based on those in
current practice. |
Note |
Based on the
Scheduler Attribute Document, this working group has the
object to define the architecture of a grid scheduler. This
definition shall serve as basis for a first implementation of
such a grid scheduler. Specifically, we want to discuss the
interaction between a grid scheduler and other components like
the grid information system, local resource management systems
and network management systems. Additionally, we want to
address the relationship between hardware resource management
and data management. In this session we present a draft for a
working group charter and a rough first draft of a document
describing this architecture. |
Name |
Laura McGinnis |
Email |
lfm@psc.edu |
WebSite |
www.mcs.anl.gov/~jms/ggf-sched/GGF5/ |
BOF: Large-site
Authentication, Authorization, and Accounting (AAA)
Requirements
Date/Time |
Tue, Jul 23,
2002: 7:30pm - 8:30pm |
Location |
Pentland
Auditorium (600) |
Area |
Security |
Description |
BOF to
discuss the implications of utilizing the Grid Security
Infratructure (GSI) on locally-defined requirements for AAA. |
Note |
Tentative
agenda: Intro and scope 5 min US Site-AAA Status 15
min EU Site-AAA Status 15 min Technical Presentation 15
min Conclusions and followup 10
min |
Name |
Dane Skow |
Email |
mailto:dane@fnal.gov |
WebSite |
http://www.ppdg.net/pa/ppdg-pa/siteaa/ |
BOF: Common
Information Model (CIM) WG
Date/Time |
Tue, Jul 23,
2002: 6:00pm - 7:00pm |
Location |
Tinto (200) |
Area |
Performance and Information Services |
Description |
The BoF will
evaluate whether CIM can be used as a common Grid Information
model which would replace the work on Grid Object
Specification activity. A common information model is needed
as base of interoperability between the object definitions of
the different Grid Working Groups. |
Note |
Early
attempts in the Grid Information Services area to define a
common grid object specification scheme (GOS) that mainly was
based on the LDAP and the XML paradigm have currently lost
momentum. The relational database paradigm in a countermove
found increasingly interest based on assumptions e.g. that it
could better map relations between objects and that it
could better cope with the increasing number of data
objects. But the new relational approach also creates new
problems that already had been solved in the older approach,
like distributedness of the data and access control via the
net.
The Common Information Model (CIM), defined by the
industrial consortium DMTF (www.dmtf.org) establishes a data
model that is aplicable to all entities that exist in the
frame of network computing (user, service, computer, policy,
etc.) . CIM does not only define attributes (and methods) of
these objects (called classes) but also different kind of
relationships between classes, that can be devided into three
groups: 1.) hierarchical inheritance, 2.) association and 3.)
aggregation. With these any number of different relations can
be defined, e.g. the relationship between policy-conditions
and policy-actions. Besides this flexibility, CIM also defines
a number of common standardized objects, like user, service,
computer, that are needed also in grid computing. CIM is
increasingly being used in commercial information system
products. The whole model can be mapped by the LDAP data
model, thus LDAP can be used to set up an information system
based on CIM.
This BoF wants to evaluate whether a new
CIM-LDAP approach for establishing a common Grid
information Service infrastructure would on one hand solve the
problems that led to the move to the relational approach (e.g.
better mapping of relations) without having to re-solve the
problems that already had been solved in the older approach
(e.g. distributednes). |
Name |
Peter Gietz,
Michael Helm |
Email |
peter.gietz@daasi.de,
helm@fionn.es.net |
WebSite |
http://www.daasi.de/wgs/CIMGIS/ |
BOF: Production Grid Management WG
Date/Time |
Tue, Jul 23,
2002: 7:30pm - 8:30pm |
Location |
Ochil 1 (50) |
Area |
Applications
& Models |
Description |
led by Tony
Lisotta |
Note |
Purpose: To explore interest in establishing a
working group that will serve as a forum to establish best
practices and standards for the administration of grids. This
BOF will bring together grid managers and administrators to
discuss issues and problems that have been encountered in
making the grid technologies all work together, the problems
and concerns that are likely to be encountered, technical as
well as non-technical, when moving a grid testbed to the
production environment. A successful outcome of this BOF will
be the establishment of a working group in this area. The
draft charter is below.
Working group charter: The purpose of this working group
is to bring together grid administrators to share experiences
and develop best practice and standards documents. This group
will explore new paradigms in supporting grids that aspire to
become large-scale grids with a large user/application base,
but will not exclude the small grid efforts. In addition, this
forum can be used to discuss problems and concerns that are
likely to be encountered, technical as well as non-technical,
when moving a grid testbed to the production environment. This
working group should be closely linked with the grid user
services group. Proposed areas to investigate within this
working group include : self-configuring grid, developing a
standard way to describe a system and its configuration, tools
that can be linked to the MDS to learn the location and
configuration of a system etc..
Draft Agenda for the first BoF: - Presentation NASA's
experience with deploying the IPG. The grid administrator
perspective DoD's Experience (not confirmed
yet) Sandia's Experience (not confirmed yet) - Open
discussion: Experiences with taking software developed in
the Grid Community and + deploying it in a production
environment. + using it to manage this environment +
using it to support users + using to keep management
satisfied - Other topics to address in this working
group - Analyzing the grid constitution and applying it to
your grid - Keeping supporting software in sync across your
grid - Non-technical issues: It takes more than just
technology - Working Group + Draft Charter for a Working
Group for comments + Volunteers?
Target
Audience Grid Administrators Grid Managers Managers
(Governance of the Grid) |
Name |
Tony Lisotta |
Email |
lisotta@nas.nasa.gov |
WebSite |
| | |