GRIDtoday Logo Altair

DAILY NEWS AND INFORMATION FOR THE GLOBAL GRID COMMUNITY /

( Previous Article )    ( Table of Contents )   

Special Features:

eBusiness HELPS DRIVE StorageTek GRID MOVE
By Derrick Harris, Editor

Here is the interview GRIDtoday conducted with Rajkumar Buyya of the University of Melbourne, and recent recipient of the StorageTek Fellowship of Grid Computing; comments from Charles Milligan, StorageTek fellow, follow.


GRIDtoday: I must say, my interest was piqued when I heard of this story -- what does a storage company want with Grid computing? Will StorageTek focus on Grid storage, or are they looking to expand into the development of Grid applications?

RAJKUMAR BUYYA: The next generation of scientific experiments and studies, popularly called as e-Science, is carried out by communities of researchers from different organizations that span national and international boundaries. Some well-known scientific experiments of this nature include the CERN-led ATLAS and CMS experiments and the KEK-led Belle experiment. These experiments involve geographically distributed and heterogeneous resources such as computational resources, scientific instruments, databases and applications. The data in these experiments is usually massive and distributed across numerous institutions for various reasons including, the inherent distribution of data sources; large-scale storage and computational requirements; to ensure high-availability and fault-tolerance of data; and caching to provide faster access. The users in such complex environments should able to carry out analysis of the data generated by the experiments by transparently accessing distributed datasets and computational resources.

Building, managing, and operating such distributed data-intensive Grids as a utility computing environment for both eScience and eBusiness applications is not only challenging, but also presents a large business opportunity for industries. I believe that this is what triggered StorageTek, a leading company with products in mass storage systems, to look into Grid computing.

Gt: Alright, now on to the discussion of money, and $125,000 is a lot of it. Does StorageTek expect that you meet any specific requirements?

BUYYA: The collaboration between StorageTek and the Gridbus Project has emerged out of our complementary interests and expertise in various aspects of Grid computing, one of which is the development of Grid economy ecosystem for autonomic management of distributed data and storage systems. Therefore, some of our research efforts are being scaled up toward meeting industrial requirements.

Gt: Speaking of Gridbus, how do you think your receiving the this fellowship will affect the work done with and the image of the Gridbus Project, and of the University of Melbourne as a whole?

BUYYA: Our collaboration with StorageTek provides an opportunity for members of the GRIDS Laboratory and its flagship Gridbus Project to work with the world's leading minds in mass storage systems. It enhances and scales up our research efforts to develop innovative Grid technologies that are designed to meet real world requirements. It also gives me a bit of extra time to dream up about the creation of disruptive Internet and Grid technologies and applications.

Receiving a StorageTek Fellowship award supporting a position at a rank equivalent to an industry-sponsored Chair of Grid Computing in a U.S. university is a great honor and recognition for both the Gridbus Project and the University of Melbourne. At the international level, one can easily view this as strengthening Australia's position as a leading destination for those in search of a place for research and innovation, and not just for holidays!

Gt: How is everything going with Gridbus?

BUYYA: The Gridbus Project has been doing very well in many respects. To name just a few: 1) our collaboration with international industries and developers of eScience and eBusiness applications is rapidly growing; 2) we continue to attract top class research scholars and Ph.D. candidates from all over the world; and 3) this year alone we have received two Discovery Project grants from the Australian Research Council.

Gt: You mentioned eBusiness, do you foresee Grid playing a major role in the business world in the near future? Do you think your work with StorageTek will directly affect Grid's adoption in the enterprise?

BUYYA: At this moment, Grid computing holds great promise in research and scientific world. Although there are several deployments of enterprise Grid systems such as Oracle10g and Sun Grid Engine, a significant amount of research, development and innovation is required for global Grids to play a major role in the business world as utility computing platforms. Our collaborative effort with StorageTek and many other international industrial and academic organizations will certainly enhance further adoption of Grid computing in enterprises.

Gt: Well, thanks for the time, and I just want ask if you have anything else to add. Feel free to do so.

BUYYA: In addition to our Grid efforts, the research work of fellow researchers from all over the world, just to name a few: 1) professor David Abramson from Monash University, Australia (who was my Ph.D. thesis advisor); 2) professors Jack Dongarra and Ian Foster from the United States; 3) professors David Walker and Mark Baker from the United Kingdom; and 4) professor Satoshi Matsuoka from Japan, who is playing a major role in enhancing the widespread adoption of Grid computing.


GRIDtoday also was able to speak to StorageTek Fellow Charles Milligan about what inspired the StorageTek Fellowship of Grid Computing, and what factors led to the decision to award the position to Buyya. Here is what he had to say:

CHARLES MILLIGAN: We had a number of goals last fall with respect to our Grid research efforts. We found several opportunities to address each of the following goals at various costs and at varying levels of creativity. The goals included:

  • Identify exactly what can (if anything) must be done to basic StorageTek products for them to be "Grid-enabled."

  • Coordinate our Grid research here with at least one university effort.

  • Extend our research reach by engaging a group of external researchers and graduate students to complement some aspect of our research.

  • Link our Grid effort with at least one international entity.

The position we sponsored at the University of Melbourne was ideal in that it addressed the last three goals. The intent is that we sponsor Dr. Buyya's position to enhance some of his research efforts particularly those addressing problems of interest to StorageTek. We selected Dr. Buyya primarily because: his research includes economic aspects of Grid management that we have not found elsewhere; he has entered his work in international Grid competitions and been in the top finishers; and he has developed Grid technologies and demonstrated their use in driving emerging eScience and eBusiness applications.

We identified the position as a StorageTek sponsored Chair at the university. However, the university pointed out that a "Chair" position is a department head (like a vice president position) and Dr. Buyya was more like a first line manager position, which the university calls a "Fellow." So, they selected the title of "Fellow of Grid Computing" and then tacked "StorageTek" on the front to acknowledge our contribution to the university. This is in no way related to the "StorageTek Fellow" position within StorageTek. This may be a one year sponsorship or ongoing -- that really depends on the value that StorageTek get out of the work this year. That is a decision that will be made in Q4 after we see the results.

( Top of Page )

( Previous Article )    ( Table of Contents )