Main Topics | Schedule | Speakers | Types of presentation | Titles (tentative) | Download |
Diner | Sunday Nov. 21st | Radio Maria |
|
| |
Workshop Day 1 (Auditorium) | Monday Nov. 22cd |
|
|
| |
Welcome and Introduction | 08:30 | Franck Cappello, INRIA & UIUC, France and Thom dunning, NCSA, USA | Background | Workshop details |
|
Post PetaScale and Exascale Systems, chair: Franck Cappello | 08:45 | Mitsuhisa Sato, U. Tsukuba, Japan | Trends in HPC | ||
| 09:15 | Marc Snir, UIUC, USA | Trends in HPC | ||
| 09:45 | Wen Mei Wu, UIUC, USA | Trends in HPC | ||
| 10:15 | Arun Rodrigues, Sandia, USA | Trends in HPC | ||
| 10:45 | Break |
|
|
|
Post Petascale Applications and System Software chair: Marc Snir | 11:15 | Pete Beckman, ANL, USA | Trends in HPC | Exascale Sofware Center | |
| 11:45 | Michael Norman, SDSC, USA | Trends in HPC | ||
| 12:15 | Eric Bohm, UIUC, USA | Trends in HPC | ||
| 12:45 | Lunch |
|
|
|
|
|
|
|
|
|
BLUE WATERS , chair Bill Gropp | 14:00 | Bill Kramer, NCSA, USA | Overview | Blue Waters: A Super-System to Explore the Expanse and Depth of 21st Century Science | |
Collaborations on System Software | 14:30 | Ana Gainaru, NCSA, USA | Early Results | ||
Steve Gottlieb | 15:00 | Esteban Menese, UIUC, USA | Early Results | Clustering Message Passing Applications to Enhance Fault Tolerance Protocols | |
| 15:30 | Thomas Ropars, INRIA, France | Results | Latest Progresses on Rollback-Recovery Protocols for Send-Deterministic Applications | |
| 16:00 | Break |
|
|
|
Collaborations on System Software, chair: Bill Kramer | 16:30 | Leonardo Bautista, Titech, Japan | Results/International collaboration with Japan | Transparent low-overhead checkpoint for GPU-accelerated clusters | |
| 17:00 | Gabriel Antoniu, INRIA/IRISA, France | Results | Concurrency-optimized I/O for visualizing HPC simulations: An Approach Using Dedicated I/O cores | |
| 17:30 | Mathias Jacquelin, INRIA/ENS Lyon | Results | ||
| 18:00 | Olivier Richard, Joseph Emeras, INRIA/U. Grenoble, France | Early Results | Studying the RJMS, applications and File System triptych: a first step toward experimental approach | |
Diner | 19:30 | Gould's |
| ||
|
|
|
|
|
|
Workshop Day 2 (Auditorium) | Tuesday Nov. 23rd |
|
|
|
|
|
|
|
|
|
|
Collaborations on System Software, chair: Raymond Namyst | 08:30 | Torsten Hoefler, NCSA, USA | Potential collaboration | ||
| 09:00 | Frederic Viven, INRIA/ENS Lyon, France | Potential collaboration | ||
Collaborations on Programming models, | 09:30 | Thierry Gautier | Potential collaboration | On the cost of managing data flow dependencies for parallel programming | |
| 10:00 | Laercio Pilla, INRIA/U. Grenoble, France | Early Results | Charm++ on NUMA Platforms: the impact of SMP Optimizations and a NUMA-aware Load Balancing | |
| 10:30 | Break |
|
|
|
chair: Sanjay Kale | 11:00 | Raymon Namyst, INRIA/U. Bordeaux, France | Potential collaboration | Bridging the gap between runtime systems and programming languages on heterogeneous GPU clusters | |
| 11:30 | Brian Amedo, INRIA/U. Nice, France | Potential collaboration | ||
| 12:00 | Christian Perez, INRIA/ENS Lyon, France | Early Results | ||
| 12:30 | Lunch |
|
|
|
Collaborations on Numerical Algorithms and Libraries, chair Mitsuhisa Sato | 14:00 | Luke Olson, UIUC, USA | Early Results | ||
| 14:30 | Simplice Donfac, INRIA/U. Paris Sud, France | Early Results | Improving data locality in communication avoiding LU and QR factorizations | |
| 15:00 | Desiré Nuentsa, INRIA/IRISA, France | Early Results | Parallel Implementation of deflated GMRES in the PETSc package | |
| 15:30 | Sebastien Fourestier, INRIA/U. Bordeaux, France | Early Results | ||
| 16:00 | Break |
|
|
|
chair: Luke Olson | 16:15 | Marc Baboulin, INRIA, U. Paris Sud, France | Early Results | Accelerating linear algebra computations with hybrid GPU-multicore systems | |
| 16:45 | Daisuke Takahashi, U. Tsukuba, Japan | Results/International collaboration with Japan | ||
| 17:15 | Alex Yee, UIUC, USA | Early Results | A Single-Transpose implementation of the Distributed out-of-order 3D-FFT | |
| 17:35 | Jeongnim Kim, NCSA, USA | Early Results | ||
Diner | 19:30 | Escobar's |
|
| |
|
|
|
|
|
|
Workshop Day 3 (Auditorium) | Wednesday Nov 24th |
|
|
|
|
|
|
|
|
|
|
Break out sessions introduction | 8:30 | Cappello, Snir | Overview | Objectives of Break-out, expected results |
|
Topics |
| Participants | Other NCSA participants |
|
|
Break out session 1 | 9:00-10:15 |
|
|
|
|
Routing, topology mapping, scheduling, perf. modeling |
| Snir, Hoefler, Vivien, Gautier, Kale, Namyst, Méhaut, Bohm, Pilla, Amedo, Perez, Baboulin |
| Room 1030 | |
Resilience |
| Kramer, Cappello, Gainaru, Ropars, Menese, Bautista, Antoniu, Richard, Fourestier, Jacquelin |
| Room 1040 | |
Libraries |
| Olson, Désiré, Simplice, |
| Room 1104 |
|
| 10:15 | Break |
|
|
|
Break out session 2 | 10:30-11:45 |
|
|
|
|
Programing models / GPU |
| Kale, Méhaut, Namyst, Wu, Amedo, Perez, Bohm, Pilla, Baboulin, Fourestier, Gautier |
| Room 1030 |
|
I/O |
| Snir, Viven, Jaquelin, Antoniu, Richard, Kramer, Gainaru, Ropars |
| Room 1040 | |
3D-FFT |
| Cappello, Takahashi, Yee, Jeongnim, Hoefler |
| Room 1104 | |
Break out session report | 12:00 | Speakers: Snir, Cappello, Kramer, Kale, Olson |
| Auditorium |
|
Closing | 12:30 | Cappello, Snir |
| Auditorium |
|
| 13:00 | Lunch |
|
|
|
Diner | 19:00 | Buttitta's |
|
|
...
In a High Performance Computing infrastructure, it is particularly difficult to master the architecture as a whole. With the physical infrastructure, the platform management software and the users' applications, understanding the global behavior and diagnosing problems is quite challenging. And it is even more true in a petascale context with thousands of compute nodes to manage and a high occupation rate of the resources. A global study of the platform will thus consider the Resource and Job Management System (RJMS), the File System and the Applications triptych as a whole. Studying their behavior is complicated because it means having some knowledge of the applications requirements in terms of physical resources and access to the File System. In this presentation, we propose a first step toward an experimental approach that mix the use of Jobs Workloads patterns and File System access patterns that, once combined, will give a full set of jobs behaviors. These synthetic jobs will then be used to test and benchmark infrastructure, considering the RJMS and the File System.
| Anchor | ||||
|---|---|---|---|---|
|
Torsten Hoefler, NCSA
Application Performance Modeling on Petascale and Beyond
...