Workshop on Enabling Data-Intensive Computing: from Systems to Applications


July 30-31, 2009

Pittsburgh, PA



Text Box: About the Workshop
 Scientific and sociological challenges are driving computations of unprecedented sizes that generate a flood of data  which must be manipulated, aggregated and analyzed. These data collections, which today range from many terabytes through Petabytes, must be accessed efficiently, made available to broad communities, and ideally, made interoperable with each other.
Computing with very large amount of data presents many challenges. Tremendous increases in computing and networking capabilities that have not been met with equal advances in storage and I/O bandwidths, requiring new software architectures to address data volumes that greatly exceed physical memory. Numerous ideas and frameworks have emerged. Some of these have attracted considerable attention, whereas others offer considerable potential but are still unexplored. Working at the extremes, hardware/systems vendors have devised innovative solutions to coping with large data, and application developers have evolved myriad strategies to mitigate insufficient I/O bandwidth.
These challenges cannot be met without active collaboration between hardware and systems software architects, data-intensive middleware developers, and algorithm and application developers. The goal of this NSF-funded workshop is to bring together these researchers, with their very different backgrounds and expertise, to understand each otherís requirements and constraints, to assess the state of the art, to identify gaps in current and planned designs where further research is required, and to develop joint strategies for enabling data-intensive computing.
Text Box: Dates and Scope
The workshop will be held on July 30-31, 2009 at the University of Pittsburgh, Pittsburgh PA.
The first day of the workshop will consist of invited talks by leading researchers on various aspects of data intensive computing, addressing: (1) architectures, (2) software frameworks and middleware, and (3) algorithms and applications. The initial list of speakers includes:
∑ John Abowd - Cornell University
∑ David Andersen - Carnegie Mellon University
∑ Magda Balazinska - The University of Washington
∑ Roger Barga - Microsoft Research
∑ Scott Brandt - The University of California at Santa Cruz
∑ Steve Elbert - Pacific Northwest National Labs
∑ Mootaz Elnozahy - International Business Machines
∑ Ian Foster - Argonne National labs
∑ Geoffrey Fox -   Indiana University
∑ Dave O'Hallaron - Intel Research
∑ Michael Wood-Vasey - University of Pittsburgh
∑ Mazin Yousef - The University of Arizona
∑ Taieb Znati - The National Science Foundation
The second day will consist of short presentations followed by break-out sessions to define new research directions and identify the collaborations needed to meet the challenges in this new and exciting area. A report will be developed from the workshop to summarize our findings. 

Organizing Committee

 Randy Bryant - Carnegie Mellon University

 Phillip Gibbons - Intel Research Pittsburgh

 Rami Melhem (Chair) - University of Pittsburgh

 Nick Nystrom - Pittsburgh Supercomputing Center

 Narayana Tummala - Google Pittsburgh


To contact us:

                        Please send email to:

Sponsored by: