BEGIN:VCALENDAR VERSION:2.0 PRODID:icalendar-ruby CALSCALE:GREGORIAN BEGIN:VEVENT DTSTAMP:20260706T061245Z UID:cee8577f-8ed5-4dbe-9091-81e1aa3f8ac8 DTSTART:20180423T070000Z DTEND:20180427T150000Z DESCRIPTION:             \n\nSponsor\n\n\n\nGoals\n\nThis worksho p organized by VI-HPS\, LRZ (new VI-HPS member since April 2018) and IT4In novations as a PRACE training event will:\n\ngive an overview of the VI-HP S programming tools suite\n explain the functionality of individual tools\ , and how to use them effectively\n offer hands-on experience and expert a ssistance using the tools\nTo foster the Czech-German collaboration in hig h performance computing\, a contingent of places has been reserved for par ticipants from the Czech Republic.\n\nProgramme Overview\n\nThe detailed p rogram will be available on the VI-HPS training web site.\n\nPresentations and hands-on sessions are planned on the following topics\n\nSetting up\, welcome and introduction\n Score-P instrumentation and measurement\n Scal asca automated trace analysis\n TAU performance system\n Vampir interactiv e trace analysis\n Periscope/PTF automated performance analysis and optimi sation\n Extra-P automated performance modeling\n Paraver/Extrae/Dimemas t race analysis and performance prediction\n [k]cachegrind cache utilisation analysis\n MAQAO performance analysis &\; optimisation\n MUST runtime error detection for MPI\n ARCHER runtime error detection for OpenMP\n JUBE script-based workflow execution environment\n ... and potentially others to be added\nA brief overview of the capabilities of these and associated tools is provided in the VI-HPS Tools Guide.\n\nThe workshop will be held in English and run from 09:00 to not later than 18:00 each day\, with brea ks for lunch and refreshments. For participants from public research insti tutions in PRACE countries\, the course fee is sponsored through the PRACE PATC program. All participants are responsible for their own travel and a ccommodation.\n\nA social event for participant and instructor networking is planned for the evening on Tuesday 24 April\, consisting of a guided to ur of the Weihenstephan Brewery sponsored by Megware followed by a self-p aid dinner at the brewery restaurant.\n\nClassroom capacity is limited\, t herefore priority may be given to applicants with parallel codes already r unning on the workshop computer system (CoolMUC-3)\, and those bringing co des from similar Xeon Phi x86 cluster systems to work on. Participants are therefore encouraged to prepare their own MPI\, OpenMP and hybrid MPI+Ope nMP parallel application codes for analysis.\n\n\n\n \n\nPicture: Partici pants of the 21st VI-HPS Tuning Workshop at LRZ in April 2016.\n\nProgramm e in Detail (provisional)\n\nDay 1:\n Monday 23 April\n 08:30\n (reg istration &\; set-up of course accounts on workshop computers)\n 09:00 \n Welcome\n Introduction to VI-HPS &\; overview of tools\n Intr oduction to parallel performance engineering\n CoolMUC-3 computer syste m and software environment [Volker Weinberg\, LRZ]\n Building and runni ng NPB/BT-MZ on CoolMUC-3 [Ilya Zhukov\, JSC]\n \n 10:30\n (break)\n  \n 11:00\n mpiP lightweight MPI profiling [Martin Schulz\, TUM]\n mpiP hands-on exercises\n MAQAO performance analysis tools [Cédric Va lensi\, Emmanuel Oseret &\; Salah Ibn Amar\, UVSQ]\n MAQAO hands-on exercises\n \n 12:30\n (lunch)\n 13:30\n Hands-on coaching to appl y tools to analyze participants' own code(s).\n 17:30\n Review of day a nd schedule for remainder of workshop\n 18:00\n (adjourn)\n  \n Day  2:\n Tuesday 24 April\n 09:00\n Score-P instrumentation &\; meas urement toolset [Ronny Tschüter\, TUD]\n Score-P hands-on exercises\n CUBE profile explorer hands-on exercises [Ilya Zhukov\, JSC]\n \n 10: 30\n (break)\n 11:00\n Score-P analysis scoring &\; measurement fi ltering  [R. Tschüter\, TUD]\n Measuring hardware counters and other m etrics \n Extra-P automated performance modeling [Sergei Shudler\, TUD armstadt]\n Extra-P hands-on exercises\n \n 12:30\n (lunch)\n 13: 30\n Hands-on coaching to apply tools to analyze participants' own code( s).\n 17:30\n Review of day and schedule for remainder of workshop\n 1 8:00\n (adjourn)\n 19:30\n Social event: Guided tour of Weihenstephan Brewery and dinner\n  \n Day 3:\n Wednesday 25 April\n 09:00\n S calasca automated trace analysis [Ilya Zhukov\, JSC]\n Scalasca hands-on exercises\n Vampir interactive trace analysis [Matthias Weber\, TUDres den]\n Vampir hands-on exercises\n \n 10:30\n (break)\n  \n 11 :00\n Paraver tracing tools suite [Judit Giménez &\; Lau Mercadal\, BSC]\n Paraver hands-on exercises\n \n 12:30\n (lunch)\n 13:30\n Hands-on coaching to apply tools to analyze participants' own code(s).\n 17:30\n Review of day and schedule for remainder of workshop\n 18:00\n (adjourn)\n  \n Day 4:\n Thursday 26 April\n 09:00\n JUBE work flow execution environment [Thomas Breuer\, JSC]\n JUBE hands-on exercis es\n Periscope Tuning Framework [Robert Mijakovic\, TUM]\n Periscope hands-on exercises\n \n 10:30\n (break)\n  \n 11:00\n TAU perf ormance system [Sameer Shende\, UOregon]\n TAU hands-on exercises\n \n 12:30\n (lunch)\n 13:30\n Hands-on coaching to apply tools to analy ze participants' own code(s).\n 17:30\n Review of day and schedule for remainder of workshop\n 18:00\n (adjourn)\n  \n Day 5:\n Friday 2 7 April\n 09:00\n MUST MPI runtime error detection [Joachim Protze\, RW TH]\n MUST hands-on exercises\n ARCHER OpenMP runtime error detection [Joachim Protze\, RWTH]\n ARCHER hands-on exercises\n \n 10:30\n (break)\n 11:00\n Kcachegrind cache analysis [Josef Weidendorfer\, LRZ] \n Kcachegrind hands-on exercises\n Review\n \n 12:30\n (lunch)\ n 13:30\n Hands-on coaching to apply tools to analyze participants' own code(s).\n 17:00\n (adjourn) \n \n \n\nHardware and Software Platfor ms\n\nCoolMUC-3: KNL-based x86 Linux cluster system:\n\n148 compute nodes each with single Intel Xeon Phi 7210-F 'Knights Landing' MIC processors (1 .3GHz\, 64 cores per processor\, 4 hardware threads per core) and 96GB RAM and 16GB HBM\n cluster modes: quad\, snc4\, a2a\n memory modes: flat\, ca che\, hybrid\n network: Intel OmniPath interconnect\n parallel filesystem: GPFS (SCRATCH &\; WORK)\n software: SLES12-based GNU/Linux\, Intel MPI \; Intel\, GCC and other compilers\; SLURM batchsystem\nThe local HPC syst em CoolMUC-3 is the primary platform for the workshop and will be used for the hands-on exercises. Course accounts will be provided during the works hop to participants without existing accounts. Other systems where up-to-d ate versions of the tools are installed can also be used when preferred\, though support may be limited and participants are expected to already pos sess user accounts on non-local systems. Regardless of whichever systems t hey intend to use\, participants should be familiar with the relevant proc edures for compiling and running their parallel applications (via batch qu eues where appropriate).\n\nRegistration\n\nVia https://events.prace-ri.eu /event/648/registration/register \n\nLocation\, Transport and Accommodatio n\n\nThe workshop will be held in Leibniz Rechenzentrum on the university campus outside Garching bei München\, approximately 25 min north from the city centre of Munich. The U-bahn line U6 (station: Garching-Forschungsze ntrum) provides direct connection from the campus area to both Munich and Garching.Getting to/from LRZ\n\nIt is recommended to choose a hotel in Gar ching or Munich city centre and use the U-bahn to reach LRZ.Accommodation in GarchingAccommodation in Munich\n\nhttps://events.prace-ri.eu/event/648 / SUMMARY:VI-HPS Tuning Workshop @ LRZ URL;VALUE=URI:https://events.prace-ri.eu/event/648/ END:VEVENT END:VCALENDAR