Big Red II Pre-Launch Workshop

Home » Cyberinfrastructure at IU » CI Training » Big Red II Pre-Launch Workshop

Big Red II Pre-Launch Workshop

The registration has closed. Please contact athota@iu.edu if you would like to attend the workshop.

Where: Cyberinfrastructure Building (CIB) Atrium (2709 East 10th Street, Bloomington, IN 47408 -- Google Maps view)

When: 9 AM – 5PM , April 9, 2013

Who should attend: Anyone who wishes to learn about the HPC services supported by IU, including HPC capabilities, hardware support, and software support, focusing on Big Red II, the newest IU supercomputer.

Abstract: The launch day for Indiana University's newest and fastest supercomputer, Big Red II, is fast approaching. We are conducting a day-long workshop to showcase the utility and capability of Big Red II on April 9th to users at IU. Big Red II is a Cray supercomputer with 1020 compute nodes, including 676 nodes with GPUs.

Topics will include:

Overview of the Big Red II system
Why GPUs are important in HPC
The New Data Capacitor
Performance Analysis with Vampir Tools

Speakers will include specialists from Cray, Nvidia, Data Direct Networks (DDN), Indiana University and Dresden University of Technology.

The registration has closed. Please contact athota@iu.edu if you would like to attend the workshop.

The Big Red II Pre-Launch Workshop is sponsored by NVIDIA and DataDirect Networks.

Attendance limit for this event is 75 people.

Agenda:
9:00 to 9:30 AM	Opening remarks and overview	Robert Henschel (Scientific Applications and Performance Tuning, IU)
9:30 to 10:15 AM	Cray, Inc. Architectures	Dave Strenski (Cray)
10:15 to 10:30 AM	Break
10:30 to 11:15 AM	Cray/Big Red II Overview	Dave Strenski (Cray)
11:15 to 12:00 PM	GPU computing 101: Why are GPUs important in HPC?	Jonathan Bentz (Nvidia)
12:00 to 1:00 PM	Lunch
1:00 to 2:00 PM	Survey of applications already ported to GPUs.	Jonathan Bentz (Nvidia)
2:00 to 2:30 PM	The New Data Capacitor	Steve Simms (High Performance File systems, IU)
2:30 to 3:00 PM	Break
3:00 to 4:00 PM	Data Capacitor Overview	Keith Miller (DDN)
4:00 to 5:00 PM	Performance Analysis Tools: Vampir, VampirTrace and Score-P	Joseph Schuchart (ZIH)

Speaker Bios:
	Dave Strenski (Cray) Dave Strenski is an Application Analyst for Cray Inc., a company that designs and manufactures high performance computers. Prior to Cray, Dave held a variety of technical positions at several computer and research organizations. Dave holds degrees in Surveying, Civil and Mechanical Engineering. His publications include works in the areas of parallel computing, numerical consistency, genomic data searching algorithms, reconfigurable computing, and other computer related topics, and was issued a patent on a meshing algorithm for threaded fasteners. As a hobby, Dave plays with solar power helping Ypsilanti residents design and install photovoltaic systems around town. Dave has given his solar presentation to over 3000 people. The presentation can be found at SolarYpsi.org along with other topical solar information.
	Joseph Schuchart (Dresden University of Technology) Joseph Schuchart is a Computer Scientist from the University of Technology Dresden, Germany, where he was working on improving the scalability of the VampirTrace tracing infrastructure. He is currently employed at the Oak Ridge National Laboratory (ORNL) to maintain the Vampir tools installation and support scientists in analyzing parallel codes on different platforms using Vampir/VampirTrace and other performance analysis tools. Current projects include the analysis of several early science applications on Titan, including CESM, LAMMPS, LSMS-WL, and NRDF.
	Jonathan Bentz (Nvidia) Dr. Jonathan Bentz is a Solutions Architect with NVIDIA, focusing on Higher Education and Research customers. In this role he works as a technical resource to customers and OEMs to support and enable adoption of GPU computing. He delivers GPU training such as workshops to train users and help raise awareness of GPU computing. He also works with ISV and customer applications to assist in optimization for GPUs through the use of benchmarking and targeted code development efforts. Prior to NVIDIA, Jonathan worked for Cray as a software engineer where he developed and optimized high performance scientific libraries such as BLAS, LAPACK, and FFT specifically for the Cray platform. Jonathan obtained his Ph.D. in physical chemistry and his MS in computer science from Iowa State University.
	Keith Miller (WW Technical Director for HPC and Life Sciences, DataDirect Networks, Inc.) Mr. Miller has been in the HPC business for almost 20 years, he previouslyworked at SGI for 12 years where he served in a number of support roles. MrMiller joined DDN in 2006 where he has been associated with many of thelargest and fastest HPC filesystems in the world. He currently serves asthe WW Technical Director for HPC and Life Sciences where he manages thepre-sales team.
	Stephen Simms (Manager, High Performance File Systems) Stephen Simms is the Data Capacitor (DC) project lead and manager of the High Performance File Systems group. He spearheads IU’s development of the Lustre file system, and is the university’s representative to OpenSFS. As the DC manager, Simms is responsible for two different filesystems, the local Data Capacitor, and the WAN Data Capacitor (DC-WAN).
	Robert Henschel (Manager, Scientific Applications and Performance Tuning) Robert Henschel leads a team of senior developers that create, optimize and maintain software for IU's high performance computing systems. The Scientific Applications and Performance Tuning group provides support for researchers at Indiana University and users of XSEDE to enable efficient use of IU's HPC systems. He is also representing Indiana University in the High Performance Group (HPG) at Standard Performance Evaluation Corporation (SPEC). In this role he leads IU's benchmarking efforts.

UITS Research Technologies