MCS Division researchers play major role in SC13

August 28, 2013

The International Conference for High Performance Computing, Networking, Storage and Analytics—better known simply as SC—has long been at the forefront in gathering the world’s leaders in supercomputing. Researchers in Argonne’s Mathematics and Computer Science (MCS) Division are playing a major role in this year’s event in November in Denver, Colorado.

Marc Snir  delivered the IEEE Computer Society Seymour Computer Engineering Award Lecture.

One paper coauthored by MCS researchers has been selected as a finalist for the Gordon Bell Prize:

  • HACC: Extreme Scaling and Performance Across Diverse ArchitecturesSalman Habib, Vitali A. Morozov, Nicholas Frontiere, Hal Finkel, Adrian Pope, Katrin Heitmann, Kalyan Kumaran, Venkat Vishwanath, Tom Peterka, Joseph A. Insley, David Daniel, Patricia Fasel, and Zarija Lukic

The ACM Gordon Bell Prize is awarded each year to recognize outstanding achievement in high-performance computing, with particular emphasis on rewarding innovation in its use in applications in science, engineering, and large-scale data analytics.

Seven other papers coauthored by MCS researchers, with topics ranging from cloud computing to autotuning to exascale application programming interfaces, have been accepted for SC13’s highly competitive Technical Papers Program.

  • Optimization of Cloud Task Processing with Checkpoint-Restart Mechanism – Sheng Di, Yves Robert, Frederic Vivien, Derrick Kondo, Cho-Li Wang, Franck Cappello
  • Characterization and Modeling of PIDX Parallel I/O for Performance Optimization – Sidharth Kumar, Avishek Saha, Venkatram Vishwanath, Philip Carns, John Schmidt, Robert Latham, Giorgio Scorzelli, Hemanth Kolla, Robert Ross, Jackie Chen, Michael E. Papka, Ray Grout, Valerio Pascucci
  • Taming Parallel I/O Complexity with Auto-Tuning – Babak Behzad, Huong Vu Thanh Luu, Joseph Huchette, Surendra Byna, Prabhat, Ruth Aydt, Quincey Koziol, Marc Snir
  • Parallelizing the Execution of Sequential Scripts – Zhao Zhang, Daniel S. Katz, Timothy G. Armstrong, Justin M. Wozniak, Ian Foster
  • SPBC: Leveraging the Characteristics of MPI HPC Applications for Scalable Checkpointing – Thomas Ropars, Tatiana Martsinkevich, Amina Guermouche, André Schiper, Franck Cappello
  • Insights for Exascale IO APIs from Building a Petascale IO API – Jay Lofstead, Robert Ross
  • SDQuery DSI: Integrating Data Management Support with a Wide Area Data Transfer Protocol – Yu Su, Yi Wang, Gagan Agrawal, Rajkumar Kettimuthu

Other contributions include organizing three tutorials:

  • Advanced MPI Programming – Pavan Balaji, James Dinan, Torsten Hoefler, Rajeev Thakur
  • Globus Online and the Science DMZ as Scalable Research Data Management Infrastructure for HPC Facilities  – Steve Tuecke, Raj Kettimuthu, Eli Dart, Vas Vasiliadis
  • Parallel I/O in Practice – Robert Latham, Robert Ross, Brent Welch, Katie Antypas

 three workshops:

  • 8th Parallel Data Storage Workshop – Robert B. Ross, Garth Gibson, John Bent
  • 6th Workshop on Many-Task Computing on Clouds, Grids, and Supercomputers (MTAGS) – Ioan Raicu, Ian Foster, Yong Zhao, Justin Wozniak
  • 8th Workshop on Ultrascale Visualization – Kwan-Liu Ma, Venkatram Vishwanath, Hongfeng Yu

and two BOFs:

  • G8 Extreme Scale and Big Data Program: Progress on Current Exascale Projects and Future Funding Opportunities – William Tang, Franck Cappello
  • MPICH: A High-Performance Open-Source MPI Implementation – Rajeev Thakur, Pavan Balaji, Rusty Lusk

In addition, MCS staff moderated two panels:

  • Exascale Runtime Systems Pavan Balaji, Laxmikant Kale, Kathy Yelick, Bronis de Supinski, Vivek Sarkar, Thomas Sterling
  • Fault Tolerance/Resilience at Petascale/Exascale: Is it really critical? Are solutions necessarily disruptive? – Franck Cappello, Marc Snir, Bronis De Supinski, Al Geist, John Daly, Ana Gainaru, Satoshi Matsuoka

Also of note is that Franck Cappello is chairing the selection of the new Test of Time Award. Beginning this year, this award will be presented annually to a single paper, selected from the SC conference proceedings of 10-25 years ago.

Two MCS Division members and their colleague  also presented a demonstration in the Scientific Visualization Showcase:

  • Secrets of the Dark Universe: Simulationg the Sky on the Blue Gene/Q, the Outer Rim Simulation – Joseph A. Insley, Salman Habib, katrin Heitmann

and in the Network Research Exhibition:

  • Enhancement of Globus GridFTP over SmartNIC User-Programmable 10GigE NIC – Rajkumar Kettimuthu, Gerald Sabin

Poster presentations by MCS members included the following:

  • Framework for Optimizing Power, Energy, and Performance – Prasanna Balaprakash, Stefan Wild
  • The Heuristic Static Load-Balancing Algorithm Applied to CESM – Yuri Alessev, Sheri Mickelson, Sven Leyffer, Robert Jacob
  • Real-Time Stochastic Optimization of Complex Energy Systems on High Performance Comptuers – Cosmin Petra, Mihai Anitescu
  • Building Scalable Data Management and Analysis Infrastructure for Metagenomics – Wei Tang, Jared Wilkening, Jared Bischof, Wofganag Gerfach, Andreas Wilke, Narayan Desai, Folker Meyer
  • A Portable Approach for Repeatabilty Testing – Quan Pham, Tanu Malik, Ian Foster
  • Optimizing the Barnes-Hut Algorithm for Multicore Clusters – Junchao Zang, Babak Behzad, Marc Snir

MCS Division staff also presented papers at several workshops:

1st Workship on Sustainable Software for Science: Practice and Experiences

  • Package Management Practices – Mark C. Miller, Lori Diachin, Satish Balay, Lois Curfman McInnes, Barry Smith
  • Accurately Citing Software and Algorithms Used in Publications – Matthew Knepley, Jed Brown, Lois Curfman McInnes, Barry Smith
  • Software as a Service as a path to software sustainability –  Ian Foster, Vas Vasiliadis, Steven Tuecke
  • Run-Time Extensibility: Anything Less Is Unsustainable – Jed Brown, Matthew Knepley, Barry Smith
  • Software Abstractions and Methodologies for HPC Simulation Codes on Future Architecturees – Anshu Dubey, S. Brandt, R. Brower, M. Giles, P. Hovland, M. Snir, R. Thakur, B. Norris, D. Q. Lamb, F. Loeffer, B. O'Shea,, C. Rebbi

4th Iinternational Workshop on Performance MOdeling, Benchmarking and Simulation of HPC Systems

  • Multiobjective Optimization of HPC Kernels for Performance, Power, and Energy – Prasanna Balaprakesh, Stefan Wild, Ananta Tiwari

WORKS 2013: 8th Workshop on Workflows in Support of Large-Scale ScienceRavi Madduri

Exascale MPIPavan Balaji


For further information, see the SC13 website.