18th VI-HPS Tuning Workshop (UGA, Grenoble, France)
Date
Monday 18th - Friday 22th May, 2015.
Location and How to Get There
The workshop will take place in Tour IRMA Salle 2 du collège doctoral of Laboratoire Jean Kuntzmann (LJK), 51 rue des Mathématiques, Campus de Saint Martin d'Hères, Université Grenoble Alpes, 38041 Grenoble, France.
Organizing Institutions
Goals
This workshop is organized by VI-HPS for the French PRACE Advanced Training Centre Maison de la Simulation and Maison de la modélisation et de la Simulation, Nanosciences et Environnement de Grenoble to:
- give an overview of the VI-HPS programming tools suite
- explain the functionality of individual tools, and how to use them effectively
- offer hands-on experience and expert assistance using the tools
Programme Overview
Presentations and hands-on sessions are on the following topics:
- Setting up, welcome and introduction
- MAQAO performance analysis & optimisation
- Score-P instrumentation and measurement
- Scalasca automated trace analysis
- Vampir interactive trace analysis
- TAU performance system
- Measurement & analysis of heterogeneous HPC systems using accelerators
A brief overview of the capabilities of these and associated tools is provided in the VI-HPS Tools Guide.
The workshop will be held in English and run from 09:00 to not later than 18:00 each day, with breaks for lunch and refreshments. There is no fee for participation, however, participants are responsible for their own travel and accommodation.
Classroom capacity is limited, therefore priority will be given to applicants with MPI, OpenMP and hybrid OpenMP+MPI parallel codes already running on the workshop computer systems, and those bringing codes from similar systems to work on.
Preliminary programme
| Day 1: | Monday 18 May | |
| 09:30 | (registration & set-up of
course accounts on workshop computers)
[Optional] Individual preparation of participants' own codes. |
|
| 12:30 |
(lunch) | |
| 14:00 |
Welcome [Marc Schlütter, JSC]
| |
| 15:30 | (break) | |
| 16:00 |
Lab setup | |
| 17:30 | (adjourn) | |
|
|
||
| Day 2: | Tuesday 19 May | |
| 09:00 |
|
|
| 10:30 | (break) |
|
| 11:00 |
MAQAO performance analysis & optimisation [A. Charif-Rubial, UVSQ] |
|
| 12:30 | (lunch) | |
| 14:00 | Hands-on coaching to apply tools to analyse participants' own code(s). | |
| 17:00 | Review of day and schedule for remainder of workshop | |
| 17:30 | (adjourn) | |
|
|
||
| Day 3: | Wednesday 20 May | |
| 09:00 |
Score-P instrumentation & measurement [Bert Wesarg, TU Dresden]
|
|
| 10:30 | (break) | |
| 11:00 |
Score-P analysis scoring & measurement filtering
|
|
| 12:30 | (lunch) | |
| 14:00 | Hands-on coaching to apply tools to analyse participants' own code(s). | |
| 17:00 | Review of day and schedule for remainder of workshop | |
| 17:30 | (adjourn) | |
|
|
||
| Day 4: | Thursday 21 May | |
| 09:00 |
Scalasca automated trace analysis [Marc Schlütter, JSC] |
|
| 10:30 | (break) |
|
| 11:00 |
Vampir interactive trace analysis [Bert Wesarg, TU Dresden] |
|
| 12:30 | (lunch) | |
| 14:00 | Hands-on coaching to apply tools to analyse participants' own code(s). | |
| 17:00 | Review of day and schedule for remainder of workshop | |
| 17:30 | (adjourn) | |
|
|
||
| Day 5: | Friday 22 May | |
| 09:00 |
TAU performance system [Sameer Shende, U Oregon] |
|
| 10:30 | (break) | |
| 11:00 |
Concluding session |
|
| 12:30 | (lunch) | |
| 14:00 | Hands-on coaching to apply tools to analyse participants' own code(s). | |
| 15:30 | (break) | |
| 16:00 | (adjourn or continue with work) |
Hardware and Software Platforms
Froggy: this compute cluster of the Grenoble University HPC mésocentre CIMENT will be used for the hands-on exercises with accounts provided during the workshop:
- 190 Bull B710 compute nodes comprised of Intel Xeon E5-2670 Sandybridge (2.60GHz, dual-socket, 8 cores per socket) and 64 or 128 GB shared memory
- 7 Bull R421-E3 GPU nodes with dual-socket Intel Xeon E5-2670 Sandybridge and Nvidia Kepler K20 GPU, 32 GB memory per node
- Infiniband FDR network
- 2 file systems: Bull HA-NFS /home 30 TB; Bull Lustre 88 TB /scratch
- software: Bullx Linux (RHEL6), Intel & GNU compilers, Bullx/Open MPI & Intel MPI, OAR batch scheduler, ...
The local system Froggy is the primary platform for the hands-on part of the workshop. Participants with applications using CUDA (with MPI and/or OpenMP) are encouraged to analyse performance on the GPU nodes. Other systems where up-to-date versions of the tools are installed can also be used when preferred, though support may be limited. Participants are expected to already possess user accounts on non-local systems they intend to use, and should be familiar with the procedures for compiling and running parallel applications on those systems.
Registration
Registration is via the PRACE training portal. Note: the number of participants is limited, and preference will be given to those bringing parallel application(s) to analyse and tune as part of the workshop.
Contact
Tuning Workshop Series
Brian Wylie
Jülich Supercomputing Centre
Phone: +49 2461 61-6589
Email: b.wylie@fz-juelich.de
Local Arrangements
Michel Kern
Maison de la Simulation
Phone: +33 1 69 08 58 32
Email: michel.kern@inria.fr
Laurence Viry
Laboratoire Jean Kuntzmann
Phone: +33 4 76 63 59 11
Email: laurence.viry@imag.fr
Sponsors
