
[Image credit: New Orleans Skyline reflected in the Mississippi River © Ed Metz | Dreamstime.com]

Happening at UT's booth (#2925) ...
- Opening night gala with Dr. Jack Dongarra's talk and UT's traditional Big Orange ice cream social
- XALT—A new tool for efficient tracking of users' jobs and environments on a cluster
- SGI tech talk—Accelerating analytics with In-memory solutions
- Other leading-edge tech talks
- Information on Intel IPCCs
- HPC Starter Kit giveaways
- Participation in the SC14 Broader Engagement scavenger hunt
When members of the high-performance computing (HPC) community from all over the world converge in New Orleans, Nov. 16–21, for the SC14 supercomputing conference, the University of Tennessee (UT) will be there, engaging in the unique meeting of researchers, scientists, engineers, educators, students, programmers, system administrators, and developers.
The conference theme, HPC Matters, emphasizes the importance of supercomputing in virtually every aspect of daily life, from drug discovery to personalized medicine, manufacturing, public safety, and finance, to myriad realms of scientific research—all for the betterment of humanity. UT will share some of the ways in which it is making a difference on that forefront as well as moving toward new horizons in supercomputing.
Leading-edge Tech Talks
On Monday, Nov. 17, at 7:30 p.m., the UT booth (#2925) will continue its tradition of being a focal point for excitement during the conference’s opening night when the audience will enjoy Big Orange ice cream while hearing a talk from Ken Kennedy-award-winning University Distinguished Professor Dr. Jack Dongarra of UT. In his talk, he will unveil the Top500 list of the world’s most powerful computers and offer expert insight into the current and future directions of HPC.

Dr. Jack Dongarra, Ken Kennedy-award-winning university distinguished professor at the University of Tennessee (UT), will present a talk on the Top500 and HPC history and future directions during the opening gala for SC14 on Monday, Nov. 17, at 7:30 p.m. in the UT booth (#2925). [Image credit (SC12, Salt Lake City): Christal Yost, National Institute for Computational Sciences]
A new HPC job-tracking tool is the subject of a talk in UT’s booth at 10 a.m. the next day, Tuesday, Nov. 18, from Dr. Mark Fahey, deputy director of the National Institute for Computational Sciences (NICS). He will speak about XALT, a mechanism for efficiently tracking user codes and environments on a computer cluster. XALT provides a census of libraries and applications and automatically filters user issues, yielding exactly the type of job tracking information that most computing centers want or need. XALT was developed through a collaborative effort of NICS and the Texas Advanced Computing Center (TACC).
Also highlighting UT’s tech talks is a contribution from SGI's Brian Freed, who will speak on the subject of accelerating analytics with in-memory solutions. Following is the complete speaker schedule.
Tech Talk Schedule |
||
Time | Speaker | Topic |
---|---|---|
Monday, November 17 | ||
7:30 p.m. | Dr. Jack Dongarra |
An Overview of High-performance Computing Today and a Look to the Future |
Tuesday, November 18 | ||
10 a.m. | Dr. Mark Fahey Deputy director of NICS |
Understanding User-level Activity with XALT |
11 a.m. | Michael Campfield NICS infrastructure team lead for the HPC Operations group and XSEDE Operations group manager |
Serving up Virtualization Two Different Ways (with a Side of Hash Tags) |
2 p.m. | George Bosilca, Piotr Luszczek, and Asim Yarkhan |
Distributed Computing at Extreme Scales • Modern Software Stack for Numerical Linear Algebra • PAPI: Performance Application Programming Interface |
4 p.m. | Pavel Shamis |
Recent Advances in OpenSHMEM and UCCS |
Wednesday, November 18 | ||
10 a.m. | George Ostrouchov Senior research scientist at Oak Ridge National Laboratory and professor of statistics at UT |
pbdR: A Sustainable Path for Scalable Statistical Computing |
11 a.m. | Brian Freed |
Accelerating Analytics with In-memory Solutions |
2 p.m. | Drew Schmidt |
Rapid Web Portal Creation with R and Shiny |
4 p.m. | Tabitha K. Samuel NICS software team lead for the HPC Operations group and XSEDE software testing and deployment manager |
UNICORE: A Middleware Tool that Enables Campus Bridging and Workflow |
Thursday, November 19 | ||
10 a.m. | Richard Knepper |
XSEDE Campus Bridging |
Student Cluster Competition
This is UT's second year participating in the SC 14 Student Cluster Competition. The group of competitors from UT for this go-around is called Team VIBE.
Tennessee Supercomputing

The Innovative Computing Laboratory (ICL) at the University of Tennessee aspires to be a world leader in enabling technologies and software for scientific computing. In addition to participating in the UT booth tech-talk speaker series again this year, ICL also will have the latest information on hand about its products, which visitors can take with them. Learn more about ICL here.

The University of Tennessee–Oak Ridge National Laboratory (ORNL) Joint Institute for Computational Sciences advances scientific discovery and state-of-the-art engineering, and furthers knowledge of computational modeling and simulation. The institute's ability to work with researchers and data scientists across the nation has set JICS on a trajectory to also become a highly valuable regional and state resource. JICS' expert consultancy combined with HPC, cluster, and shared-memory computing technical resources, offers opportunities for partnerships for those in a data-driven industry or a proposal-reliant research setting. These partnerships can solve the toughest problems in science, engineering, or Big Data society faces in the 21st century. In addition to traditional scientific domains, JICS is optimally positioned to very effectively collaborate with those in the social, behavioral, and economics communities as well. Learn more about JICS here.

The National Institute for Computational Sciences (NICS), one of the nation’s leading academic supercomputing centers, had the distinction of deploying and managing Kraken, the first academic supercomputer to break the petaflop barrier of a quadrillion floating-point operations per second. NICS is a major partner in the National Science Foundation's eXtreme Science and Engineering Discovery Environment, known as XSEDE. Learn more about NICS here.