• Subscribe

Things are moving very fast, says 'father of the grid' Ian Foster

Ian Foster, director of the Computation Institute at theUniversity of Chicago & Argonne National Laboratory, often referred to asone of the 'fathers of the grid', is the general chair and moderator of next week's 8th IEEE International Conference on eScience (eScience 2012). iSGTW spoke to him about his hopes for the event:

A photograph of Ian Foster.
Ian Foster. Image courtesy Peter Kiar. Homepage image courtesy Wikimedia Commons.

What are the key areas of focus of the eScience2012 event and how will you directly be involved in the program?

I always enjoy the eScience conference because of its interdisciplinary mix of users and developers. eScience 2012 is particularly exciting because things are moving very fast at present; for example, the simultaneous emergence of large-scale cloud computing, big data, 100 gigabit-per-second networks, and petascale computers - Chicago and Illinois are at the center of many of these developments. The eScience 2012 program covers all of these topics and more. To highlight just four sessions of interest:

(1) Charlie Catlett's panel on 'Designing and operating cities using open data' engages global leaders in the use of open data to understand urban systems, and ultimately to make cities more livable and sustainable.

(2) Carole Goble's keynote talk on 'The reality of reproducibility of computational science' speaks on how we may achieve the ideal of reproducible science in an era of ever-more-automated computational processes.

(3) Lenny Smith's keynote talk on 'Uncertainty and predictability in climate forecasts' addresses the fundamental question of when and how we can put faith in the outputs of computational models.

(4) Gerhard Klimeck's keynote talk on 'Mythbusting knowledge transfer with nanoHUB.org' celebrates an early success story in the large-scale use of cloud computing for science, namely the nanoHUB system, which does online simulations for nanotechnology.

The rest of the technical itinerary that program chairs Dan Katz, of the University of Chicago and Argonne National Laboratory, US, and Heinz Stockinger, from the SIB Swiss Institute of Bioinformatics, Switzerland, have put together is also excellent. I look forward to an immensely stimulating, if exhausting, week.

eScience 2012 has a number of commercial supporters (Microsoft Research, EMC2 and Cray, the supercomputer company). How is commercial support helping you address research challenges from a computational perspective?

Microsoft, EMC2, and Cray are all major players in eScience, although from quite different perspectives. Microsoft is expanding its efforts in cloud computing in a big way, via its Azure project, which will be highlighted in the Microsoft workshop on Monday and Tuesday. EMC2 is a major supplier of storage technologies and via its acquisitions of companies, such as VMware and Greenplum, is an increasingly important player in cloud and big data. Cray is one of the biggest suppliers of high-performance computers, including both the University of Chicago's Beagle and the NCSA's Blue Waters. So, each of our supporters is an important contributor to the work that we do.

How much coverage will there be about cloud computing's impact on eScience? What are the main challenges of integrating cloud technologies with high-performance computing today?

eScience has always been concerned with how on-demand computing can accelerate discovery: a topic that has become mainstream as commercial cloud providers have emerged. The big issue now is how to leverage commercial on-demand computing in scientific research. I see this as requiring innovation at multiple levels, from research methods to software platforms and even mundane issues such as accounting and billing.

Homepage image banner of the eScience 2012 event in Chicago.
This year, the eScience event will take place in the
Windy City of Chicago. Image courtesy eScience

Many talks at the event cover workflows. What aspects of these sessions do you think attendees should watch out for and why?

The numerous talks on workflow-related topics are an indication of the growing importance of automation in eScience. With data volumes growing exponentially, scientists need to automate if they are to keep up. Take a look, for example, at the research papers by Karastoyanova et al, and Nguyen and Abramson, on integrating user interaction into workflow systems.

In your opinion, what are the quick wins in eScience to enable researchers to take better advantage of their data? How are these being addressed at the conference?

I find it interesting to see more papers on data mining at eScience this year, and also papers on large-scale infrastructures for data federation. For example, the Earth System Grid Federation links some 20 sites worldwide with many petabytes of climate simulation data. We've come a long way since the first eScience conference!

Do you have any concluding remarks for readers?

I look forward to seeing many of you in Chicago. Putting together a conference of this quality and diversity is a big job. Meeting you at the conference makes it all worthwhile.

The first eScience conference took place in Melbourne, Australia, in 2005. This year's event is the 8th in the popular series and will run from 8 to 12 October 2012. It will be held at the Hyatt Regency hotel in Chicago, Illinois, US. You can see the event program here or register to the conference through this link.

iSGTW is a media partner of eScience 2012.

Join the conversation

Do you have story ideas or something to contribute? Let us know!

Copyright © 2019 Science Node ™  |  Privacy Notice  |  Sitemap

Disclaimer: While Science Node ™ does its best to provide complete and up-to-date information, it does not warrant that the information is error-free and disclaims all liability with respect to results from the use of the information.


We encourage you to republish this article online and in print, it’s free under our creative commons attribution license, but please follow some simple guidelines:
  1. You have to credit our authors.
  2. You have to credit ScienceNode.org — where possible include our logo with a link back to the original article.
  3. You can simply run the first few lines of the article and then add: “Read the full article on ScienceNode.org” containing a link back to the original article.
  4. The easiest way to get the article on your site is to embed the code below.