Virtual Institute – High Productivity Supercomputing

10th VI-HPS Tuning Workshop (LRZ, Germany)

Date

October 16-19, 2012

Location

Garching-bei-München
LRZ / TUM

Organizing Institutions

LRZ PRACE

Goals

This hands-on workshop organized by VI-HPS for Gauss Centre for Supercomputing PRACE Advanced Training Centre will:

  • 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

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, meals and accommodation.

Programme

Day 1 Tuesday 16 October
09:00 (registration & set-up of course accounts on workshop computers)
[Optional] Individual preparation of participants' own codes.
12:00 (lunch on your own)
13:00 (registration)
13:30 Welcome & Introduction to VI-HPS [Geimer, JSC]
  • Introduction to parallel performance analysis [Oleynik, TUM]
  • 15:00 (break)
    15:30 Overview of VI-HPS tools [Geimer, JSC]
    16:30 Lab setup
  • Computer systems and software environment [LRZ]
  • Building and running NPB-MZ-MPI/BT-MZ [Geimer, JSC]
  • 17:30 (adjourn)
    19:00 Social dinner (Gasthof Neuwirt, Garching) sponsored by ParTec
    Day 2 Wednesday 17 October
    09:00 Score-P instrumentation & measurement toolset [Philippen, JSC]
  • Score-P hands-on exercises
  • CUBE hands-on exercises [Böhme, GRS]
  • 10:30 (break)
    11:00 Score-P analysis scoring & measurement filtering [Geimer, JSC]
  • Measuring hardware counters and other metrics [Tschüter, TUD-ZIH]
  • TAU ParaProf / PerfExplorer hands-on exercises [Spear, UOregon]
  • 12:30 (lunch on your own)
    13:30 Hands-on coaching to apply tools to analyze participants' own code(s).
    17:00 Review of day and schedule for remainder of workshop
    17:30 (adjourn)
    Day 3 Thursday 18 October
    09:00 Trace collection and analysis
  • Vampir interactive trace analysis [Tschüter, TUD-ZIH]
    Vampir hands-on exercises
  • Scalasca automated trace analysis [Geimer, JSC]
    Scalasca hands-on exercises
  • 10:30 (break)
    11:00 Additional instrumentation and measurement options
  • Periscope online performance analysis [Petkov, TUM]
    Periscope hands-on exercises
  • TAU performance system [Spear, UOregon]
    TAU/PDT hands-on exercises
  • 12:30 (lunch on your own)
    13:30 Hands-on coaching to apply tools to analyze participants' own code(s).
    17:00 Review of day and schedule for remainder of workshop
    17:30 (adjourn)
    Day 4 Friday 19 October
    09:00 KCachegrind toolset [Weidendorfer, TUM]
  • KCachegrind hands-on tutorial exercises
  • 10:30 (break)
    11:00 MAQAO/ECR tools [Krammer, UVSQ]
  • MAQAO/ECR hands-on tutorial exercises
  • 12:30 (lunch on your own)
    13:30 Hands-on coaching to apply tools to analyze participants' own code(s).
    15:00 (break)
    15:30 (adjourn or continue with work to 16:30)

    Classroom capacity is limited, therefore priority will be given to applicants with parallel codes already running on the workshop computer systems, and those bringing codes from similar systems to work on. Participants are therefore encouraged to prepare their own MPI, OpenMP and hybrid MPI+OpenMP parallel application codes for analysis.

    VI-HPS Tools

    • KCachegrind is a free cache-utilization visualization tool developed by TUM.
    • MAQAO free modular assembly code quality analyzer and optimizer framework developed by UVSQ.
    • PAPI is a free library interfacing to hardware performance counters developed by UTK-ICL and used by many tools.
    • Periscope is an automatic performance analysis tool using a distributed online search for performance bottlenecks being developed by TUM.
    • Scalasca is an open-source toolset developed by JSC & GRS to analyze the performance behaviour of MPI & OpenMP parallel applications and automatically identify inefficiencies.
    • TAU is a performance system for measurement and analysis of parallel programs written in Fortran, C, C++, Java & Python developed by the University of Oregon.
    • Vampir is a commercial framework and graphical analysis tool developed by TUD-ZIH to display and analyze trace files.
    • Score-P, OPARI2, OTF2 & CUBE4 are components of a community parallel performance instrumentation, measurement & analysis infrastructure.

    Hardware and Software Platforms

    The local HPC system is expected to be the primary platforms for the workshop, with priority for improved job turnaround and local system support. Course accounts will be provided for those who need them.

    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 the systems.


    Local arrangements & registration

    Register via the LRZ registration form choosing course HHPS1W12:
    http://www.lrz.de/services/compute/courses/#VIHPS

    Although unclear from the Teilnahmebedingungen on the LRZ registration form, participants from PRACE partner countries are especially welcome.

    The workshop will be held in Leibniz Rechenzentrum on the campus of Technische Universität München outside Garching bei München, approximately 25 min north from the city centre of Munich. The U-bahn line U6 (station: Garching-Forschungszentrum) provides direct connection from the campus area to both Munich and Garching.
    Getting to/from LRZ

    It is recommended to choose a hotel in Garching or Munich city centre and use the U-bahn to reach LRZ.
    Accomodation in Garching
    Accomodation in Munich

    Contact

    Brian Wylie (JSC), phone +49 2461 61-6589
    Ventsislav Petkov (TUM), phone +49 89-289-17680