|
09/12/2006, 4:45 PM - 5:30 PM
Speaker:
Wolfgang Gentzsch, Duke University.
Many of the early regional, national, and community grid initiatives were built around the Globus Toolkit, mainly due to the growing world-wide community of developers and users, the funding from US government and industry to further develop the technology, and the leadership and evangelization of the Globus team. After four major releases, today, Globus technology is mature enough and ready to be deployed by the "early majority" in research and enterprises. For this new generation of users, technology and infrastructure are often of secondary importance: they want to solve a problem in computational sciences, engineering or business.
With our presentation, we would like to help this community to better understand, build, manage and operate grids, based on the experience of the early adopters and on case studies and lessons learned from selected grid projects. For this purpose, we are currently analyzing major grid projects and infrastructures around the world, like the US TeraGrid, Naregi in Japan, UK e-Science Program, European EGEE, and the German D-Grid initiative. We will summarize Lessons Learned, draw conclusions and provide recommendations for those who intend to build national or community grids in the near future.
Our research so far is based on several interviews with major representatives from the UK e-Science initiative, EGEE, and D-Grid, and next step will be with Naregi and TeraGrid. As an example, one of the earliest, most funded, and therefore most important ones is the UK e-Science Initiative. Major e-Science projects have been studied and key representatives interviewed from six e-Science Centers. Major focus of our research and of the interviews was on applications and strategic direction, government and industry funding, national and international cooperation, and on strengths and weaknesses of the grid projects. The interviewees answered questions about their role in the project(s), successful and less successful projects, sustainability of the results, integration of individual results into the bigger picture, available funding, the future of e-Science, international collaboration, and commercial services. As a result, we have compiled a list of "Lessons Learned" and "Recommendations and Conclusions" which may serve others to successfully plan, implement, operate and fund similar grid projects in the near future. The first paper focusing on the UK e-Science initiative is available upon request.
We are currently collecting similar information from and about the other grid projects, to provide a final summary of the results and conclusions and recommendations during the presentation at GlobusWORLD.
|
