Conference for High Performance Computing and Communications

Education Program

EducationJanuary 30, 2020 Priscilla Williamson

SC2001 offers high school and middle school teachers an opportunity to learn computer modeling and simulation and the application of computational science to the science and mathematics curricula. Through a national competition, a core group of 27 teams with four teachers each has been identified and will participate in an 18-month program that starts at SC2001, continues with monthly videoconference sessions on specific computational science topics through the winter, and includes a two-week Summer Institute in July 2002. The following Education Program sponsors fund these teams:

• National Science Foundation
• Association of Computing Machinery
• IEEE Computer Society
• IEEE Foundation
• Cisco Systems, Inc.
• Compaq Computer Corporation
• High Performance Systems
• Microsoft Corporation
• National Aeronautics and Space Administration
• SBC DataComm
• SC2001 Conference
• Shodor Education Foundation, Inc.
• Wolfram Research, Inc.

At the conclusion of this program, the selected teachers will become leaders in their school systems and region, providing inspiration for a wider adoption of modeling and simulation by classroom teachers. Each team will learn state-of-the-art modeling software tools that will enable them to create new classroom modules that adhere to the national science and mathematics standards. These modules will then be placed in a repository and made publicly available.

Additional teams or individual teachers can participate in the SC2001 Conference along with the selected teams by registering for the SC2001 Education Program and attending the Education Program sessions. Full participation in the hands-on sessions will require additional participants to bring wireless laptops. These participants will learn the fundamentals of the computational science curriculum development tools and will learn how to select appropriate topics for computational science modules for classroom instruction through interaction with modeling experts and practicing computational scientists. Interested teachers can receive additional information by sending email to education@sc2001.org or by visiting the SC2001 website at http://www.sc2001.org.

Education Program Speakers

Richard Allen, Albuquerque High Performance Computing Center
Lisa Bievenue, National Center for Supercomputing Applications
Edna Gentry, University of Alabama in Huntsville
Bob Gotwals, Shodor Education Foundation
Barbara Helland, Krell Institute
Jeff Huskamp, East Carolina University
Eric Jakobsson, National Center for Supercomputing Applications
Cynthia Lanius, Rice University
Scott Lathrop, National Center for Supercomputing Applications
Ernie Marshburn, East Carolina University
Robert Panoff, Shodor Education Foundation
Helen Parke, East Carolina University
Susan Ragan, Maryland Virtual High School
Jeffrey C. Huskamp, Education Chair
East Carolina University

Lisa Bievenue, Education Co-Chair
National Center for Supercomputing Applications

Edna Gentry, Education Co-Chair
University of Alabama at Huntsville

Technical Papers

SC2001January 30, 2020 Priscilla Williamson

The technical papers component of the SC2001 program is especially strong and vibrant. Sixty papers were selected from a pool of 240 submissions covering a broad technical scope and offering a truly international perspective on high performance networking and computing. Eighty-four of the papers were student submissions, of which seventeen were accepted and six nominated for the best student paper award.

In my view, SC has come into its own with this year’s program. Several of the papers represent the unique combination of real science, novel computational methods, and in-depth performance analysis on leading-edge platforms that only comes together at SC. You will find sessions such as “Sea, Wind, and Fire,” “Groundbreaking Applications,” and “Material Science” tackling computational modeling of the ocean, atmosphere, combustion, and structures. Several of these are finalists for the Gordon Bell Prize. Clusters and Grids are well represented from hardware and software viewpoints, reflecting the emergence of these important platforms. In-depth design studies explore leading commericial architectures, novel hardware devices, and emerging interconnects. Theoretical and empirical studies investigate new parallel numerical methods, algorithms, and performance-analysis techniques. Networking and storage are explored from unusual angles, and eScience emerges in the context of new modes of interacting with computational processes and high performance architectures.

Together, these papers provide a rich treatment of the latest technical work over the broad scope of high performance networking and computing. They are complemented with workshops on selected topics and a leading Masterworks track of invited speakers.

I would like to thank the members of the Technical Papers Committee and the Program Committee, who worked so hard to bring it all together and the authors from around the world who have built such a strong program. I think you too will find it exciting.

Sessions Chronologically	Sessions Alphabetically
Architectures Scientific Computing on the Itanium™… ACM IEEE The Sun Fireplane System Interconnect… ACM IEEE Parallel Graphics and Interactivity… ACM IEEE	Algorithmic Load Balancing Large Scale Parallel Structured… ACM IEEE Parallel Interval-Newton… ACM IEEE Dynamic Load Balancing of SAMR… ACM IEEE
Software Scalability Scalable Parallel Application Launch… ACM IEEE Scaling Iregular Parallel Codes… ACM IEEE A Parallel Java Grande Benchmark… ACM IEEE	Algorithms Stable, Globally Non-iterative… ACM IEEE Stochastic Search for Signal… ACM IEEE A Hypergraph-Partitioning Approach… ACM IEEE
Communication Structures ORT—A Communication Library… ACM IEEE On-the-fly Calculation… ACM IEEE Removing the Overhead… ACM IEEE	Architectures Scientific Computing on the Itanium™… ACM IEEE The Sun Fireplane System Interconnect… ACM IEEE Parallel Graphics and Interactivity… ACM IEEE
Material Science Applications Scalable Atomistic Simulation… ACM IEEE An 8.61 Tflop/s Molecular… ACM IEEE Multi-teraflops Spin Dynamics… ACM IEEE	Communication Structures ORT—A Communication Library… ACM IEEE On-the-fly Calculation… ACM IEEE Removing the Overhead… ACM IEEE
Mesh Methods Achieving Extreme Resolution… ACM IEEE A Distributed Memory Unstructured… ACM IEEE Multilevel Algorithms for Generating… ACM IEEE	Computational Grid Applications Applying Scheduling and Tuning… ACM IEEE An Automatic Design Optimization… ACM IEEE Numerical Libraries and the Grid… ACM IEEE
Computational Grid Portals and Networks The XCAT Science Portal ACM IEEE A Jini-based Computing… ACM IEEE Efficient Network and I/O… ACM IEEE	Computational Grid Environments and Security Supporting Efficient Execution… . pdf 208K Optimisation of Component-based… . pdf 128K Adapting Globus and Kerberos… . pdf 108K
Computational Grid I/O LegionFS: A Secure and Scalable… ACM IEEE High-Performance Remote Access… ACM IEEE Gathering at the Well… ACM IEEE	Computational Grid I/O LegionFS: A Secure and Scalable… ACM IEEE High-Performance Remote Access… ACM IEEE Gathering at the Well… ACM IEEE
Algorithmic Load Balancing Large Scale Parallel Structured… ACM IEEE Parallel Interval-Newton… ACM IEEE Dynamic Load Balancing of SAMR… ACM IEEE	Computational Grid Portals and Networks The XCAT Science Portal ACM IEEE A Jini-based Computing… ACM IEEE Efficient Network and I/O… ACM IEEE
Sea, Wind, and Fire Coastal Ocean Modeling… ACM IEEE Terascale Spectral Element… ACM IEEE High Resolution Weather Modeling… ACM IEEE	Efficient Layouts for Hierarchical Memories A Ghost Cell Expansion… ACM IEEE Improving Parallel Irregular… ACM IEEE Increasing Temporal Locality… ACM IEEE
Reconfigurable Architectures Delivering Acceleration… ACM IEEE Parallel Dedicated hardware… ACM IEEE Cost Effectiveness of an Adaptable… ACM IEEE	Fast I/O MPI-IO/GPFS… ACM IEEE A Case Study in Application… ACM IEEE The Design of I/O-efficient… ACM IEEE
Groundbreaking Applications Solution of a Three-Body… ACM IEEE Parallel Implementation and… ACM IEEE Modeling of Seismic Wave… ACM IEEE	Groundbreaking Applications Solution of a Three-Body… ACM IEEE Parallel Implementation and… ACM IEEE Modeling of Seismic Wave… ACM IEEE
Information Retrieval and Transaction Processing Efficient Execution of Multiple… ACM IEEE Dynamic Page Placement… ACM IEEE Compressing Inverted Files… ACM IEEE	Information Retrieval and Transaction Processing Efficient Execution of Multiple… ACM IEEE Dynamic Page Placement… ACM IEEE Compressing Inverted Files… ACM IEEE
Performance Prediction On Using SCALEA for Performace… ACM IEEE Modeling and Detecting Performance… ACM IEEE Predictive Performance and Scalability… ACM IEEE	Material Science Applications Scalable Atomistic Simulation… ACM IEEE An 8.61 Tflop/s Molecular… ACM IEEE Multi-teraflops Spin Dynamics… ACM IEEE
Algorithms Stable, Globally Non-iterative… ACM IEEE Stochastic Search for Signal… ACM IEEE A Hypergraph-Partitioning Approach… ACM IEEE	Mesh Methods Achieving Extreme Resolution… ACM IEEE A Distributed Memory Unstructured… ACM IEEE Multilevel Algorithms for Generating… ACM IEEE
Novel Graphics and Grids Fast Matrix Multiplies… ACM IEEE Next-Generation Visual… ACM IEEE Global Static Indexing… ACM IEEE	Networking EMP: Zero-copy OS-bypass… ACM IEEE Design and Implementation of FMPL… ACM IEEE A New Routing Mechanism… ACM IEEE
Computational Grid Applications Applying Scheduling and Tuning… ACM IEEE An Automatic Design Optimization… ACM IEEE Numerical Libraries and the Grid… ACM IEEE	Novel Graphics and Grids Fast Matrix Multiplies… ACM IEEE Next-Generation Visual… ACM IEEE Global Static Indexing… ACM IEEE
Networking EMP: Zero-copy OS-bypass… ACM IEEE Design and Implementation of FMPL… ACM IEEE A New Routing Mechanism… ACM IEEE	Performance Prediction On Using SCALEA for Performace… ACM IEEE Modeling and Detecting Performance… ACM IEEE Predictive Performance and Scalability… ACM IEEE
Computational Grid Environments and Security Supporting Efficient Execution… ACM IEEE Optimisation of Component-based… ACM IEEE Adapting Globus and Kerberos… ACM IEEE	Reconfigurable Architectures Delivering Acceleration… ACM IEEE Parallel Dedicated hardware… ACM IEEE Cost Effectiveness of an Adaptable… ACM IEEE
Efficient Layouts for Hierarchical Memories A Ghost Cell Expansion… ACM IEEE Improving Parallel Irregular… ACM IEEE Increasing Temporal Locality… ACM IEEE	Sea, Wind, and Fire Coastal Ocean Modeling… ACM IEEE Terascale Spectral Element… ACM IEEE High Resolution Weather Modeling… ACM IEEE
Fast I/O MPI-IO/GPFS… ACM IEEE A Case Study in Application… ACM IEEE The Design of I/O-efficient… ACM IEEE	Software Scalability Scalable Parallel Application Launch… ACM IEEE Scaling Iregular Parallel Codes… ACM IEEE A Parallel Java Grande Benchmark… ACM IEEE

AWARDS CAP SC2001

SC2001January 30, 2020 Priscilla Williamson

HIGH-PERFORMANCE COMPUTING AND NETWORKING CONFERENCE

DENVER, CO — SC2001, the conference of high-performance networking and computing capped the most successful programs in its history by recognizing outstanding achievements and contributions in these fields.

The conference attracted nearly 200 exhibitors and over 5200 attendees, each of whom on average will spend between $3.5- and $7-million on planned purchases as a result of the week-long event.

The awards presented this afternoon honored a range of people and their accomplishments.

The primary award was the third annual IEEE Computer Society Seymour Cray Computer Engineering Award. This honor recognizes innovative contributions to high-performance computing systems that exemplify Seymour Cray’s creative spirit.

This year it was awarded to John L. Hennessy, President of Stanford University. Included is a $10,000 honorarium funded by an SGI endowment.

The Gorden Bell Prize is named after the Digital Equipment vice president. It is awarded annually at SC for the best peak computer performance, the best performance/price ratio and for a special category.

The Gordon Bell winner was the team of Toshiyuki Fukushige and Junichiro Makino, who achieved a simulation of black holes in a galactic center at a computer processing speed of 11.55 trillion floating operations per second.

The Gordon Bell Price/Performance prize went to Joon Hwang, Seung Kim and Chang Lee. Their study of impact locating on aircraft structure by low-cost cluster cost 24.6 cents/Mflops, or less than a U.S. quarter per 1-million floating operations per second.

The winner of the Gordon Bell Prize in the special category of supporting efficient execution in the heterogeneous distributed computing environments with Cactus and Globus. The winner was the team of Gabrielle Allen,Thomas Dramlitsch, Ian Foster, Nick Karonis, Matei Ripeanu, Edward Seidel and Brian Toonen.

The Best Student Paper was awarded to a computational grid application of tomography, a technique to reconstruct the three-dimensional structure of an object from a series of two-dimensional projections. The award went to a team of Shava Smallen, Henri Cazsanova and Francine Berman, and carries a $500 cash award.

Finally, the award for Best Research Poster went to Sumir Chandra, Johan Steensland and Manish Parashar.

Reconfigurable Architectures

ArchitecturesJanuary 30, 2020 Priscilla Williamson

Chair: Steve Reinhardt, Silicon Graphics Inc.

Title: Delivering Acceleration: The Potential for Increased HPC Application Performance Using Reconfigurable Logic
Authors:
David Caliga (SRC Computers, Inc)
David Peter Barker (SUPERsmith)
Abstract:
SRC Computers, Inc. has integrated adaptive computing into its SRC-6 high-end server, incorporating reconfigurable processors as peers to the microprocessors. Performance improvements resulting from reconfigurable computing can provide orders of magnitude speedups for a wide variety of algorithms. Reconfigurable logic in Field Programmable Gate Arrays (FPGAs) has shown great advantage to date in special purpose applications and specialty hardware. SRC Computers is working to bring this technology into the general purpose HPC world via an advanced system interconnect and enhanced compiler technology.
Title: Parallel Dedicated Hardware Devices for Heterogeneous Computations
Authors:
Alessandro Marongiu (CASPUR, Roma)
Paolo Palazzari (ENEA-HPCN, Roma)
Vittorio Rosato (ENEA-HPCN, Roma)
Abstract:
We describe a design methodology which allows a fast design and prototyping of dedicated hardware devices to be used in heterogeneous computations. The platforms used in heterogeneous computations consist of a general-purpose COTS architecture which hosts a dedicated hardware device; parts of the computation are mapped onto the former, parts onto the latter, in a way to improve the overall computation efficiency. We report the design and the prototyping of a FPGA-based hardware board to be used in the search of low-autocorrelation binary sequences. The circuit has been designed by using a recently developed Parallel Hardware Generator (PHG) package which produces a synthesizable VHDL code starting from the specific algorithm expressed as a System of Affine Recurrence Equations (SARE). The performance of the realized devices has been compared to those obtained on the same numerical application on several computational platforms.
Title: Cost Effectiveness of an Adaptable Computing Cluster
Authors:
Keith D. Underwood (Clemson University)
Ron R. Sass (Clemson University)
Walter B. Ligon, III (Clemson University)
Abstract:
With a focus on commodity PC systems, Beowulf clusters traditionally lack the cutting edge network architectures, memory subsystems, and processor technologies found in their more expensive supercomputer counterparts. What Beowulf clusters lack in technology, they more than make up for with their significant cost advantage over traditional supercomputers. This paper presents the cost implications of an architectural extension that adds reconfigurable computing to the network interface of Beowulf clusters. A quantitative idea of cost-effectiveness is formulated to evaluate computing technologies. Here, cost-effectiveness is considered in the context of two applications: the 2D Fast Fourier Transform (2D-FFT) and integer sorting.

« Prev 1 2

Conference for High Performance Computing and Communications

The Conference of High-Performance Computing and Networking

Awards Highlight

Research Exhibits

Will Digital Actors Replace Human Ones?

Plenaries