Titan Summit
The Oak Ridge Leadership Computing Facility (OLCF) will host a summit at Oak Ridge National Laboratory (ORNL). The summit will provide a forum for current and prospective users to meet with OLCF staff and industry vendors to discuss Titan, the next-generation leadership-class computational resource, and a significant step towards exascale computing at the OLCF. Users are encouraged to present their application plans for scaling to exascale.
Registration is required for this event. Information about travel and accommodations in the Oak Ridge area are provided below. Lunch and refreshments will be provided during the event.
Additionally, the event will be webcast. Please indicate if you intend to participate in person or via webcast on the registration form.
[tab:Overview] Summit OverviewThe purpose of this summit is to bring together OLCF staff, current and prospective users, and vendors to discuss Titan and its implications to application development. The summit is intended to be an open forum to discuss techniques, strategies, concerns, and expectations. Users are encouraged to discuss their application roadmaps to help OLCF staff provide the appropriate assistance to ensure that user’s realize their research goals on Titan.
A variety of presentations from OLCF staff and vendors about lessons learned and the current state of development tools will serve as a backdrop to enhance the discussions about user applications with respect to their path towards exascale computing.
[tab:Agenda]Agenda (subject to change):
| Monday, August 15, 2011 | ||
| 8:00-9:00 | Registration | |
| 9:00-9:15 | Welcome (Refreshments provided) | Robert Whitten |
| 9:15-9:45 | Titan Overview | Robert Whitten |
| 9:45-10:45 | Titan, path-forward to Multi-petaflop and Exaflop Computing | John Levesque |
| 10:45-11:00 | Break | |
| 11:00-12:00 | Using Cray’s Hybrid Multi-core Programming environment for porting applications to Titan | John Levesque |
| 12:00-1:00 | Working Lunch | |
| GPU: Introduction | Adam Simpson | |
| 1:00-1:30 | LAMMPS: Code Transformations in preparing for Titan | Arnold Tharrington |
| 1:30-2:00 | Preparing WL-LSMS for First Principles Thermodynamics Calculations on Accelerator and Multicore Architectures | Markus Eisenbach |
| 2:00-2:15 | Break | |
| 2:15-2:45 | Porting S3D turbulent combustion software to accelerator based systems | Ramanan Sankaran |
| 2:45-3:15 | Porting the Denovo Radiation Transport Code to Titan: Lessons Learned | Wayne Joubert |
| 3:15-5:00 | Open Forum | |
| 5:00 | Adjourn | |
| Tuesday, August 16, 2011 | ||
| 8:00-9:00 | Registration | |
| 9:00-9:15 | Welcome (Refreshments provided) | Robert Whitten |
| 9:15-10:15 | HMPP Workbench: Technical Overview | Jean-Charles Vasnier (remote) |
| 10:15-10:30 | Break | |
| 10:30-11:30 | DDT: Getting started with GPU debugging | David Lecomber (remote)
David Maples |
| 11:30-1:00 | Working Lunch | |
| GPU: Vector Add Example | Adam Simpson | |
| 1:00-2:00 | Debugging for CUDA Accelerated MPI Codes | Chris Gottbrath |
| 2:00-2:15 | Break | |
| 2:15-3:15 | Trace Based Performance Analysis at Large Scale on Jaguar and Titan | Guido Juckeland |
| 3:15-4:15 | Computing with Interlagos and Nvidia processors | Dave Norton |
| 4:15-4:30 | Break | |
| 4:30-5:00 | MAGMA: LAPACK for HPC on heterogeneous architectures | Stanimire Tomov |
| 5:00 | Adjourn | |
| Wednesday, August 17, 2011 | ||
| 8:00‐9:00 | Registration | |
| 9:00‐9:15 | Welcome (Refreshments provided) | Robert Whitten |
| 9:15-9:45 | Scalable Modeling and Simulation: An Overview of Extreme Scale Application Development at PNNL | T.P. Straatsma |
| 9:45‐10:15 | NIM: A unified model for forecasting weather and climate | Jim Rosinski |
| 10:15‐10:30 | Break | |
| 10:30‐11:00 | Computing the Dark Universe | Salman Habib (remote) |
| 11:15‐12:00 | Chimera on Titan: Exploiting new parallelism for supernova simulations | Eric Lentz |
| 12:00‐1:00 | Working Lunch | |
| GPU: Game of Life Example | Adam Simpson | |
| 1:00‐1:30 | Progress Towards Accelerating CAM-SE on Hybrid Multi-Core Systems | Rich Archibald |
| 1:30‐2:00 | Two case studies of Monte Carlo simulation on GPU | Junqi Yin |
| 2:00‐2:30 | To the ExaScale and Beyond: Statistically Motivated Linear Scaling for Protein Dynamics | TJ Lane |
| 2:30-2:45 | Break | |
| 2:45-3:45 | Quantum Monte Carlo, keeping up with the HPC evolution | Jeongnim Kim |
| 3:45-4:15 | Next generation simulation of biomolecules | Pratul Agarwal |
| 4:15‐4:45 | Titan Website | Robert Whitten |
| 4:45 | Adjourn | |
In-person Attendance
Please follow this link to ORNL’s website providing area directions: Maps and Directions
For Oak Ridge area hotels: Hotels
Detailed information about visiting ORNL Information: ORNL Visitor Information
Webcast
Participants who are unable to attend in person are encouraged to participate in the webcast of the summit. Please indicate on the registration form if you will be attending the summit in-person or via webcast. Instructions for the joining the webcast will be sent via email prior to the start of the summit.
August 15, 2011
Titan Overview – Robert Whitten
GPU: Introduction – Adam Simpson
LAMMPS: Code Transformations in preparing for Titan – Arnold Tharrington
Porting S3D turbulent combustion software to accelerator based systems – Ramanan Sankaran
Porting the Denovo Radiation Transport Code to Titan: Lessons Learned – Wayne Joubert
August 16, 2011
HMPP Workbench: Technical Overview – Jean-Charles Vasnier
DDT: Getting started with GPU debugging – David Lecomber & David Maples
GPU: Vector Add Example – Adam Simpson
Debugging for CUDA Accelerated MPI Codes – Chris Gottbrath
Trace Based Performance Analysis at Large Scale on Jaguar and Titan – Guido Juckeland
Computing with Interlagos and Nvidia processors – Dave Norton & Craig Toefler
MAGMA: LAPACK for HPC on heterogeneous architectures – Stanimire Tomov
August 17, 2011
NIM: A unified model for forecasting weather and climate – Jim Rosinski
Computing the Dark Universe – Salman Habib
Chimera on Titan: Exploiting new parallelism for supernova simulations – Eric Lentz
GPU: Game of Life Example – Adam Simpson
Progress Towards Accelerating CAM-SE on Hybrid Multi-Core Systems – Rich Archibald
Two case studies of Monte Carlo simulation on GPU – Junqi Yin
To the ExaScale and Beyond: Statistically Motivated Linear Scaling for Protein Dynamics – TJ Lane
Quantum Monte Carlo, keeping up with the HPC evolution – Jeongnim Kim
