State of the Practice
An exciting new element for SC11, the State of the Practice provides a
venue for the discussion and exchange of best practices involving
provisioning, using and improving the critical systems and services in high
performance computing, networking, and storage. SC11 recognizes many of the
community challenges that span institutions, architectures and vendors,
even within a single facility. The challenges include improving performance
at scale, large-scale system management and deployment, highly parallel
storage, and energy efficiency, to name a few. It is valuable to provide
an open exchange of best ideas and methods for dealing with the aspects of
unprecedented scale. The State of the Practice component is a formal,
peer-reviewed venue organized to address this need. Reports and
presentations will be available as a permanent record within the ACM and
IEEE CS archives. We encourage organizations and individuals that may not
routinely submit in the technical program to consider this new forum.
State of the Practice (SotP) will have three sessions per day, Tuesday through
Thursday, featuring 26 accepted reports from universities, national labs and vendors in the U.S., Asia and Europe. The inaugural session at 10:30 a.m. Tuesday, Nov. 15, will follow the Technical Program focus of "Sustained Performance" with the presentation of reports on varying aspects of performance measurement, benchmarks and metrics (seven in all).
SotP presents the opportunity for all "purveyors" and "consumers" of HPC to share and learn from one another. SotP brings together a range of multi-million dollar HPC centers and more moderately funded centers to highlight their experiences, implementations and "lessons learned" for everyone to use.
SotP is also an excellent opportunity for SC11 vendors to learn for themselves how their customers are fielding and supporting their technology products.
Each of the nine sessions are themed to highlight specific aspects of HPC:
- Performance I
- Performance II
- Perf/Viz/Deploy
- Cloud/Grids
- AICS/Security/Net
- Infra/Cooling/Mgmt
- Applications
- HPC Monitoring/Syslog (featuring five leading HPC Centers)
- HPC Health/User Support
Note: A Tech Program badge is required to attend the SotP sessions.
Questions: state-of-the-practice@info.supercomputing.org
Tuesday, November 15th
| TIME
| PRESENTATION
| SPEAKER
| LOCATION
| PLANNER
|
| 10:30AM - 11:00AM |
Deep and Wide Metrics for HPC Resource Capability and Project Usage |
David L. Hart |
TCC 202 |
   |
| 11:00AM - 11:30AM |
How to Measure Useful, Sustained Performance |
William Kramer |
TCC 202 |
|
| 11:30AM - 12:00PM |
Sustained Systems Performance Monitoring at the Department of Defense High Performance Computing Modernization Program |
Paul M. Bennett |
TCC 202 |
|
| 1:30PM - 2:00PM |
Performance Evaluations of Gyrokinetic Eulerian Code GT5D on Massively Parallel Multi-Core Platforms |
Yasuhiro Idomura, Sebastien Jolliet |
TCC 202 |
|
| 2:00PM - 2:30PM |
System-Level Monitoring of Floating-Point Performance to Improve Effective System Utilization |
Davide Del Vento, Thomas Engel, Siddhartha S. Ghosh, David L. Hart, Rory Kelly, Si Liu, Richard Valent |
TCC 202 |
|
| 2:30PM - 3:00PM |
Performance Modeling for Systematic Performance Tuning |
Torsten Hoefler, William Gropp, Marc Snir, William Kramer |
TCC 202 |
|
| 3:30PM - 4:00PM |
The NWSC Benchmarks: Using Scientific Throughput as a Measure for Supercomputer Performance |
Rory C. Kelly, Siddhartha S. Ghosh, Davide Del Vento, Si Liu, Richard A. Valent |
TCC 202 |
|
| 4:00PM - 4:30PM |
Integrating Multi-Touch in High-Resolution Display Environments |
Brandt Westing, Ben Urick, Maria Esteva, Freddy Rojas, Weijia Xu |
TCC 202 |
|
| 4:30PM - 5:00PM |
Best Practices for the Deployment and Management of Production HPC Clusters |
Karl W. Schulz, Robert McLay, William L. Barth, Tommy Minyard |
TCC 202 |
|
Wednesday, November 16th
| TIME
| PRESENTATION
| SPEAKER
| LOCATION
| PLANNER
|
| 10:30AM - 11:00AM |
Wallaby: A Scalable Semantic Configuration Service for Grids and Clouds |
William C. Benton, Robert H. Rati, Erik J. Erlandson |
TCC 202 |
|
| 11:00AM - 11:30AM |
Cloud Versus In-house Cluster: Evaluating Amazon Cluster Compute Instances for Running MPI Applications |
Yan Zhai, Mingliang Liu, Jidong Zhai, Xiaosong Ma, Wenguang Chen |
TCC 202 |
|
| 11:30AM - 12:00PM |
Qserv: A Distributed Shared-Nothing Database for the LSST Catalog |
Daniel L. Wang, Serge Monkewitz, Kian-Tat Lim, Jacek Becla |
TCC 202 |
|
| 1:30PM - 2:00PM |
Japanese National HPC Research Institute and its 10petaflops Supercomputer "K" |
Akinori Yonezawa, Tadashi Watanabe, Mitsuo Yokokawa, Mitsuhisa Sato, Kimihiko Hirao |
TCC 202 |
|
| 2:00PM - 2:30PM |
Intrusion Detection at 100G |
Scott Campbell, Jason Lee |
TCC 202 |
|
| 2:30PM - 3:00PM |
A Long-Distance InfiniBand Interconnection Between Two Clusters in Production Use |
Sabine Richling, Steffen Hau, Heinz Kredel, Hans-Günther Kruse |
TCC 202 |
|
| 3:30PM - 4:00PM |
Co-Designing HPC Systems and Facilities |
Henry M. Tufo, Michael K. Patterson, Michael Oberg, Matthew Woitaszek, Guy Cobb, Robert Strong, Jim Gutowski |
TCC 202 |
|
| 4:00PM - 4:30PM |
"Hot" for Warm Water Cooling |
Henry Coles, Michael Ellsworth, David Martinez |
TCC 202 |
|
| 4:30PM - 5:00PM |
Challenges in the Management of High-Performance Computing Centers: An Organizational Perspective |
Nicholas Berente, Jennifer Claggett |
TCC 202 |
|
Thursday, November 17th
| TIME
| PRESENTATION
| SPEAKER
| LOCATION
| PLANNER
|
| 10:30AM - 11:00AM |
A Survey of the Practice of Computational Science |
Prakash Prabhu, Thomas B. Jablin, Arun Raman, Yun Zhang, Jialu Huang, Hanjun Kim, Nick P. Johnson, Feng Liu, Soumyadeep Ghosh, Stephen Beard, Matthew Zoufaly, Taewook Oh, David Walker, David I. August |
TCC 202 |
|
| 11:00AM - 11:30AM |
Adaptive Simulation of Turbulent Flow Past a Full Car Model |
Niclas Jansson, Johan Hoffman, Murtazo Nazarov |
TCC 202 |
|
| 11:30AM - 12:00PM |
World-highest Resolution Global Atmospheric Model and Its Performance on the Earth Simulator |
Keiko Takahashi, Koji Goto, Yoshinori Tsuda, Yuki Tochihara, Ken’ichi Itakura, Satoru Okura, Kunihiko Watanabe, Hiromitsu Fuchigami, Akira Azami, Kenryo Kataumi, Jun-ichi Yamamoto, Hiroshi Takahara, Toshifumi Takei, Yoshiyuki Kubo, Yoko Isobe |
TCC 202 |
|
| 1:30PM - 2:30PM |
Challenges of HPC Monitoring |
William (Bill) E. Allcock, Randal Rheinheimer, Mike Lowe, Joshi Fullop, Evan Felix |
TCC LL4/LL5 |
|
| 2:30PM - 3:00PM |
LOGJAM: A Scalable Unified Log File Archiver |
Nicholas P. Cardo |
TCC LL4/LL5 |
|
| 3:30PM - 4:00PM |
A Toolkit for Event Analysis and Logging |
James Carey, Philip Sanders |
TCC LL2 |
|
| 4:00PM - 4:30PM |
SPOTlight on Testing - Stability, Performance, and Operational Testing of LANL HPC Clusters |
Georgia A. Pedicini, Jennifer K. Green |
TCC LL2 |
|
| 4:30PM - 5:00PM |
One Stop High Performance Computing User Support at Sandia National Laboratories |
John A. Greenfield, Lisa G. Ice, Sophia E. Corwell, Karen Haskell, Constantine Pavlakos, John P. Noe |
TCC LL2 |
|
|
