TACC BOOTH PRESENTATIONS


Wednesday, November 19, 2014

PRESENTER DETAILS

TIME

DK Panda, OSU
MVAPICH2 and MVAPICH2-MIC: Designs and Experience on Stampede

11:00AM

Jim Browne, UT Austin
Quick Start Guide for Using Intel Xeon Phis

12:00PM

Thursday, November 20, 2014

Meet and Greet Brunch

10:00AM

Tuesday, November 18, 2014

PRESENTER DETAILS

TIME

Robert McLay, TACC
Transitioning to Lmod

11:00AM

Dan Stanzione, Director, TACC
TACC Overview

2:00PM

Kelly Gaither, Director of Visualization, TACC
Vis @ TACC

3:30PM


OFFICIAL SC14 PRESENTATIONS


Sunday, November 16

Programming the Xeon Phi  

Event Type: Tutorial | Time: 8:30AM - 5:00PM | Location: Room 394

The Innovative Technology component of the recently deployed XSEDE Stampede supercomputer at TACC provides access to 8 PetaFlops of computing power in the form of the new Intel Xeon Phi Coprocessor, also known as a MIC. While the MIC is x86 based, hosts its own Linux OS, and is capable of running most user codes with little porting effort, the MIC architecture has significant features that are different from that of present x86 CPUs, and optimal performance requires an understanding of the possible execution models and basic details of the architecture. This tutorial is designed to introduce attendees to the MIC architecture in a practical manner. Multiple lectures and hands-on exercises will be used to acquaint attendees with the MIC platform and explore the different execution modes as well as parallelization and optimization through example testing and reports.

CHAIR/PRESENTER DETAILS:

Lars Koesterke - Texas Advanced Computing Center

Kent Milfeld - Texas Advanced Computing Center

Dan Stanzione - Texas Advanced Computing Center

Jerome Vienne - Texas Advanced Computing Center

Monday, November 17

VISTech 2014: Visualization Infrastructure and Systems Technology  

Event Type: Workshops | Time: 9:00AM - 5:30PM | Location: Room 294

Human perception is centered on the ability to process information contained in visible light, and our visual interface is a tremendously powerful data processor. Every day we are inundated with staggering amounts of digital data. For many types of computational research, visualization is the only viable means of extracting information and developing understanding from this data. Integrating our visual capacity with technological capabilities has tremendous potential for transformational science. We seek to explore the intersection between human perception and large-scale visual analysis through the study of visualization interfaces and interactive displays. This rich intersection includes: virtual reality systems, visualization through augmented reality, large scale visualization systems, novel visualization interfaces, high-resolution interfaces, mobile displays, and visualization display middleware. The VISTech workshop will provide a space for experts in the large-scale visualization technology field and users to come together to discuss state-of-the art technologies for visualization and visualization laboratories.

CHAIR/PRESENTER DETAILS:

Kelly Gaither - Texas Advanced Computing Center

Jason Leigh - University of Hawaii at Manoa

Falko Kuester - University of California, San Diego

Eric Wernert - Indiana University

Aditi Majumder - University of California, Irvine

Karla P. Vega - Indiana University

BE PROGRAMMING CHALLENGE  

Event Type: HPC Interconnections (BE, Undergraduates, Cluster) | Time: 3:30PM - 7:00PM | Location: Room 288/289

The Broader Engagement programming challenge is an opportunity for new programmers and seasoned programmers within the BE program to compete against each other in friendly competition. With multiple levels and multiple languages, there is something for everyone to excel at. You only need your laptop to compete, as a virtual machine will be provided with all the needed software needed. Participants can also use their own software if they prefer. Test your abilities by writing legible code in a timely manner. Top competitors will receive prizes.

CHAIR/PRESENTER DETAILS:

Kenneth Craft (Chair) - Intel Corporation

Raquell Holmes - improvscience

Ritu Arora - University of Texas at Austin

Wrangler: A New Generation of Data Intensive Cluster Computing  

Event Type: Booth Presentation | Time: 10:30AM - 11:30PM | Location: Dell Booth 1739

Experimental Infrastructures for Open Cloud Research  

Event Type: Birds of a Feather | Time: 12:15PM - 1:15PM | Location: Room 388/389/390

Cloud Computing has emerged as a critical infrastructure for scientific, enterprise, and commercial computing. To support the creation of such infrastructure, there is a need for quality testbeds for development and testing. Commercial companies create their own testbeds, but academic and government cloud researchers don't have access to them. Therefore, funding agencies around the world have or soon will construct an array of new experiment infrastructures for in cloud computing. This BOF will provide a forum for the providers of these experimental infrastructures and their potential users to come together, and form a community of testbeds.

CHAIR/PRESENTER DETAILS:

Kate Keahey - Argonne National Laboratory

Warren Smith - Texas Advanced Computing Center

Dan Stanzione - Texas Advanced Computing Center

Tuesday, November 18

A User-Friendly Approach for Tuning Parallel File Operations  

Event Type: Papers | Time: 2:00PM - 2:30PM | Location: Room 393/394/395

The Lustre file system provides high aggregated I/O bandwidth and is in widespread use throughout the HPC community. Here we report on work (1) developing a model for understanding collective parallel MPI write operations on Lustre, and (2) producing a library that optimizes parallel write performance in a user-friendly way. We note that a system's default stripe count is rarely a good choice for parallel I/O, and that performance depends on a delicate balance between the number of stripes and the actual (not requested) number of collective writers. Unfortunate combinations of these parameters may degrade performance considerably. For the programmer, however, it's all about the stripe count: an informed choice of this single parameter allows MPI to assign writers in a way that achieves near-optimal performance. We offer recommendations for those who wish to tune performance manually and describe the easy-to-use T3PIO library that manages the tuning automatically.

CHAIR/PRESENTER DETAILS:

Mark Gary - Lawrence Livermore National Laboratory

Si Liu - Texas Advanced Computing Center

Robert McLay - Texas Advanced Computing Center

John Cazes - Texas Advanced Computing Center

Doug James - Texas Advanced Computing Center

William Barth - Texas Advanced Computing Center

Wednesday, November 19

TBD  

Event Type: Booth Presentation | Time: 10:30AM - 11:30PM | Location: Dell Booth 1739

CHAIR/PRESENTER DETAILS:

Dan Stanzione - Texas Advanced Computing Center

Super-R: Supercomputing and R for Data-Intensive Analysis  

Event Type: Birds of a Feather | Time: 5:30PM - 7:00PM | Location: Room 388/389/390

R has become popular for data analysis in many areas due to its high-level expressiveness and multitude of domain-specific packages. Challenges still remain in developing methods to effectively scale R to the power of supercomputers, and in deploying and enabling access to end users. This year's BOF will focus on usability and provisioning R as an interactive high performance analysis environment within HPC resources. This BOF will consist of a set of short presentations and discussions with audience. The ultimate goal is to help build a community of users and experts interested in applying R to solve data intensive problems.

CHAIR/PRESENTER DETAILS:

Weijia Xu - Texas Advanced Computing Center

Hui Zhang - Indiana University

George Ostrouchov - Oak Ridge National Laboratory

Xeon Phi Users Group: Performance Tuning and Functional Debugging for Xeon Phi  

Event Type: Birds of a Feather | Time: 5:30PM - 7:00PM | Location: Room 286/287

This BOF will build community among those developing HPC applications for systems incorporating the Intel Xeon Phi many-core processor. Taking advantage of the processor's full capabilities requires tuning and optimizing using programming techniques and tools targeted at a combination of CPUs and the Xeon Phi Coprocessor. Threading, vectorization, memory contiguity and alignment, and data locality may be important for performance. This BOF will combine a few brief presentations sharing insights and best practices with a moderated discussion among all those in attendance. It will close with an invitation to an ongoing discussion through the Intel Xeon Phi Users Group (IXPUG).

CHAIR/PRESENTER DETAILS:

Richard Gerber - National Energy Research Scientific Computing Center

Kent Milfeld - Texas Advanced Computing Center

Chris Gottbrath - Rogue Wave Software, Inc.

Thursday, November 20

Drilling Down: Understanding User-Level Activity on Today's Supercomputers with XALT  

Event Type: Birds of a Feather | Time: 12:15PM - 1:15PM | Location: Room 386/387

Let's talk real, no-kiddin' supercomputer analytics: drilling down to the level of individual batch submissions, users, and binaries. And we're not just targeting performance: we're after everything from which libraries and functions are in demand to preventing the problems that get in the way of successful science. This BoF will bring together those with experience and interest in technologies that can provide this type of job-level insight. To this end, the proposers will show off their new tool named XALT and have a short demo that the attendees can do on their own laptops.

CHAIR/PRESENTER DETAILS:

Robert McLay - Texas Advanced Computing Center

Mark Fahey - University of Tennessee, Knoxville

Friday, November 21

HUST14: First International Workshop on HPC User Support Tools  

Event Type: Birds of a Feather | Time: 3:30AM - 12:00PM | Location: Room 297

Researchers pushing the boundaries of science and technology are an existential reason for supercomputing centres. In order to be productive, they heavily depend on HPC support teams, who in turn often struggle to adequately support the researchers.

Nevertheless, recent surveys have pointed out that there is an abundant lack of collaboration between HPC support teams all around the world, even though they are frequently facing very similar problems with respect to providing end users with the tools and services they require.

With this workshop we aim to bring together all parties involved, i.e. system administrators, user support team members, tool developers, policy makers and end users, to discuss these issues and bring forward the solutions they have come up with. As such, we want to provide a platform to present tools, share best practices and exchange ideas that help streamline HPC user support. More information: https://ugent.be/hpc/hust14.html

  • 9.00am - 9.30am: Comprehensive Resource Use Monitoring for HPC Systems with TACC Stats - Richard Evans (TACC), Bill Barth (TACC), James Browne (UT Austin), Robert Deleon (Univ. at Buffalo), Thomas Furlani (Univ. at Buffalo), Steven Gallo (Univ. at Buffalo), Matthew Jones (Univ. at Buffalo) and Abani Patra (Univ. at Buffalo)
  • 10.30am - 11.00am: User Environment Tracking and Problem Detection with XALT - Kapil Agrawal (UT Knoxville), Mark R. Fahey (NICS), Robert McLay (TACC) and Doug James (UT Austin)
  • 11.00am - 11.30am: Modern Scientific Software Management Using EasyBuild and Lmod - Markus Geimer (JSC), Kenneth Hoste (HPC-UGent) and Robert McLay (TACC)

CHAIR/PRESENTER DETAILS:

Ralph C. Bording - iVEC, University of Western Australia

Andy Georges - Ghent University

Kenneth Hoste - Ghent University