Top Banner
© 2006 Open Grid Forum Geoffrey Fox OGF 23 Barcelona GFSG Meeting June 1 2008 OGF eScience Function
8

OGF eScience Function

Feb 25, 2016

Download

Documents

talon

OGF eScience Function. Geoffrey Fox OGF 23 Barcelona GFSG Meeting June 1 2008. Some Principles and Issues. Aim at Relevance and Excellence exploiting loose flexible interactions with Standards and Enterprise - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: OGF eScience Function

© 2006 Open Grid Forum

Geoffrey FoxOGF 23 Barcelona

GFSG MeetingJune 1 2008

OGF eScience Function

Page 2: OGF eScience Function

© 2006 Open Grid Forum

Some Principles and Issues• Aim at Relevance and Excellence exploiting loose flexible

interactions with Standards and Enterprise• Few activities will lead to new standards but loose coupling of

eScience and Standards will be mutually fruitful• Many activities jointly interesting and best jointly organized by

Enterprise and eScience• Problem to address: Attract more of eScience community.

• I believe one must aim at the leaders in field as these ensure excellence. • Current attendance levels due partly to view that OGF is a

standards organization (Forum part not common perception)• But dominant problem that “Grids are stale” and in US “Grid funding”

small and “(Tera)Grid directions” unclear• Grids no longer on Gartner Hype Chart

• Important to attract more software providers

Page 3: OGF eScience Function

© 2006 Open Grid Forum3

Gartner 2006 Technology Hype Curve

Grids entering trough of disillusionment

Page 4: OGF eScience Function

© 2006 Open Grid Forum4

Gartner 2007 Technology Hype Curve

No Grids! Semantic Web 10 years outWeb Services also in declineClouds will surely be at peak of inflated expectations in 2008

Page 5: OGF eScience Function

© 2006 Open Grid Forum

General Approach

• A few key community/research groups such as GIN and (perhaps in future) Clouds

• Major part of program is event community call and special workshops organized by leadership

• Special series like software provider forum but need large attendance to work• Difficult to fit into current events; European events have

enough attendance but little time• Disseminate through web-based record and special

issues of journals. • Currently 8 submissions to general OGF 2008 special issue • Previously had special issues on portals (3), workflow, GIS

and provenance with a total of 80 published papers. • Do approximately one special issue per year

Page 6: OGF eScience Function

© 2006 Open Grid Forum

OGF Special Issue 2008Concurrency&Computation: Practice&Experience• Reliability in Grid Computing Systems• The HPC Basic Profile and SAGA: Standardizing Compute Grid

Access in the Open Grid Forum• Accessing RDF(S) Data Resources in Service-based Grid

Infrastructures • WS-Naming: Location Migration, Replication, and Failure

Transparency Support for Web Services • Profiles for Conveying the Secure Communication

Requirements of Web Services • A General Encoding Framework for Representing Network

Measurement and Topology Data• Interoperation of World-Wide Production e-Science

Infrastructures • Using Clouds to Provide Grids Higher-Levels of Abstraction and

Explicit Support for Usage Modes

Page 7: OGF eScience Function

© 2006 Open Grid Forum

Immediate Steps: Clouds • Continue current approach with a focus on clouds that is very

interesting to many from all OGF functions• This requires careful planning which is ongoing• Note “data area” is also interesting to many but mature field (like

workflow) and OGF not well regarded in this area• Clouds are new and OGF can lead and define attractive

activities that will draw people• Cloud difficulties

• Clouds did not come from Grid community• Quite a bit of competition; CCC from CRA and activities at Grid

conferences (HPDC will have a Cloud panel)• Some think Clouds are “just Grids re-hyped”• OGF does not have strong contacts with leaders of Cloud initiatives or

virtualization• Not easy to make major steps at OGF24-25 as clouds mainly US based;

OGF26 is a long time!• Activity on LinkedIn Cloud forum 100 times that on OGF cloud mailing list• Not much interest in cloud standards

Page 8: OGF eScience Function

© 2006 Open Grid Forum

Other Observations

• Attendance at eScience events high but perhaps mainly people who come for standards?

• Opportunities for better “web pages” drawing interest but • huge competition• Thought Leadership Series drew no interest from

inside or outside OGF• Reduced staffing makes harder

8