9/17/2014 | (Wednesday) | Session Chair | Speaker | Organization |
---|---|---|---|---|
9:00 | Workshop Introduction and Opening Remarks | Srinath Vadlamani | NCAR | |
9:10 | Session 1 - Common Features and Challenges of Using Multi-core Systems | Chris Kerr | NOAA | |
9:10 - 9:30 | Workflow tools for refactoring legacy code | Youngsung Kim | NCAR | |
9:30 - 9:50 | Modeling Parallelism in a Shotgun-marriage Climate Application | Hans Johansen | LBL | |
9:50 -10:20 | Morning Break | |||
10:20 - 10:40 | Efficient Implicitness: Latency-throughput and cache-vectorization tradeoffs | Jed Brown | ANL/CU | |
10:40 - 11:00 | Optimizing Weather Model Radiative Transfer Physics for the Many Integrated Core and GPGPU Architectures | John Michalakes | NOAA | |
11:00 - 11:30 | Session 1 Discussion | |||
11:30 - 1:00 | Lunch | |||
1:00 | Session 2 - Optimizing for Xeon and Xeon Phi | Ilene Carpenter | NREL | |
1:00 - 1:20 | Performance Optimization for the NERSC-8 machine: The NERSC application readiness project | Nick Wright | NERSC/LBL | |
1:20 - 1:40 | Application Readiness on NERSC-8 | Matt Cordery | NERSC/LBL | |
1:40 - 2:00 | Weather and Earth System Modeling with Intel Xeon Phi: Practice and Experience | Richard Tran Mills | Intel | |
2:00 - 2:20 | Improvements in Intel Compilers XE 2015 | Milind Girkar | Intel | |
2:20 - 2:50 | Session 2 Discussion | |||
2:50 - 3:20 | Afternoon Break | |||
3:20 | Session 3 - Porting and Optimizing for GPUs | Srinath Vadlamani | NCAR | |
3:20 - 3:40 | Cray OpenACC Implementation in Various Weather and Climate Kernels | Eric Dolven | Cray | |
3:40 - 4:00 | NVIDIA HPC Directions for Earth System Modeling | Stan Posey | NVIDIA | |
4:00 - 4:20 | Evaluation of Fortran OpenACC Compilers | Mark Govett | NOAA | |
4:20 - 4:50 | Session 3 Discussion | |||
4:50 - 5:00 | First day summary | Chris Kerr | NOAA | |
5:00 | Adjourn |
9/18/2014 | (Thursday) | |||
---|---|---|---|---|
9:00 | Session 4: Common Features and Challenges of Using Multi-core Systems | Mark Govett | NOAA | |
9:00 - 9:20 | Current State of Directive-Based Accelerated Computing | Jeff Larkin | NVIDIA | |
9:20 - 9:40 | Faster Communication for Fine-grain Computing | Jacques Middlecoff | NOAA | |
9:40 - 10:00 | The GFDL FMS: An Analysis of Subdomain Halo Imposed Scaling Limitations | Jeff Durachta | NOAA | |
10:00 -10:30 | Session 4 Discussion | |||
10:30 -11:00 | Morning Break | |||
11:00 | Session 5a: Porting and Optimizing for Intel Xeon and Xeon Phi | Bill Putman | NASA | |
11:00 - 11:20 | Progress Porting and Tuning NWP Physics Packages on Intel Xeon Phi | Tom Henderson | NOAA | |
11:20 - 11:40 | From Knights ENDGame to Gung Ho beginnings: Porting old and developing new dynamical cores in the age of accelerators | Christopher Maynard | Met Office | |
11:40 - 12:00 | The GOcean project: applying the Gung-Ho computational framework to ocean modelling | Mike Ashworth | STFC Daresbury Lab | |
12:00 - 1:10 | Lunch - on your own in the NCAR cafeteria | |||
1:10 | Session 5b: Porting and Optimizing for Intel Xeon and Xeon Phi | Bill Putman | NASA | |
1:10 - 1:30 | Current status of porting and optimization efforts for NOAA/ESRL weather models on Intel Xeon Phi architecture | Jim Rosinski | ||
1:30 - 1:50 | Progress of CESM on Stampede’s Xeon Phi system | Srinath Vadlamani | NCAR | |
1:50 - 2:20 | Multi-core efforts within the MPAS Framework with a focus on the Ocean core | Doug Jacobsen | LANL | |
2:20 - 2:50 | Session 5 discussion | |||
2:50 - 3:10 | Afternoon Break | |||
3:10 | Session 6: Optimizing for GPUs | Ilene Carpenter | NREL | |
3:10 - 3:30 | Porting the Community Atmospheric Model - Spectral Element (CAM-SE) to utilize hybrid GPU architectures | Matt Norman | ORNL | |
3:30 - 3:50 | HOMME code on hybrid architectures: performance analysis and optimization strategy | Irina Demeshko | SNL | |
3:50 - 4:20 | Progress porting WRF to GPU using OpenACC | Carl Ponder | NVIDIA | |
4:20 - 4:50 | Session 6 discussion | |||
4:50 - 5:10 | Workshop Summary and Discussion | Chris Kerr | NOAA | |
5:10 | Adjourn |