r Prttrs r t s - PowerPoint PPT Presentation

❚r❛✐♥✐♥❣ Pr❛❝t✐t✐♦♥❡rs ❢♦r ❊✛❡❝t✐✈❡ ❯s❡ ♦❢ ❍P❈ ❙②st❡♠s✿ ❊①♣❡r✐❡♥❝❡ ❢r♦♠ t❤❡ ❖✛❡r❡❞ ❈♦✉rs❡s ❚❡❛❝❤✐♥❣ ❍✐❣❤ ❚❤r♦✉❣❤♣✉t ❈♦♠♣✉t✐♥❣ ❆♥❥❛ ●❡r❜❡s ●♦❡t❤❡ ❯♥✐✈❡rs✐t② ♦❢ ❋r❛♥❦❢✉rt ❏✉❧② ✸✵✱ ✷✵✶✾

❆❝t✐✈✐t✐❡s ✵✾ / ✷✵✶✺ ✵✷ / ✷✵✶✻ ✶✵ / ✷✵✶✼ ✺ ♠♦♥t❤s ✷✵ ♠♦♥t❤s ✼ ♠♦♥t❤s st❛rt ♦❢ ♠② st❛rt ♦❢ ❝♦✉rs❡ ❚❛❧❦ ♦♥ ❡♠♣❧♦②♠❡♥t ❝♦♥tr❛❝t s❡ss✐♦♥s ❡❘❡s❡❛r❝❤❆✉str❛❧❛s✐❛ ❛t ●♦❡t❤❡ ❯♥✐✈❡rs✐t② ❈♦♥❢❡r❡♥❝❡ ✐♥ ❋r❛♥❦❢✉rt ✫ ❍❑❍▲❘ ❆✉str❛❧✐❛ ✵✺ / ✷✵✶✽ ✵✼ / ✷✵✶✾ ✶✷ / ✷✵✷✵ ✶✹ ♠♦♥t❤s ✶✼ ♠♦♥t❤s st❛rt ♠❡♠❜❡rs❤✐♣ ❛t ❲♦r❦s❤♦♣✿ ❍P❈ ❡♥❞ ♦❢ ❝♦♥tr❛❝t ❍P❈❈❋ tr❛✐♥✐♥❣✱ ❡❞✉❝❛t✐♦♥ ❛♥❞ ❞♦❝✉♠❡♥t❛t✐♦♥ ❛t ❘❘❩

❯s❡r ❋❡❡❞❜❛❝❦ Questionnaire Anja Gerbes Introductory Courses in Frankfurt Date: 14.02.2019 Evaluation of the Talk very good Which course . . . you join? very bad did would How would you evaluate. . . UNIX � � TOOLS � � . . . the content and target of the workshop? � � � � � � GIT � � actuality SHELL � � � � � � � � comprehensibility � � � � � � CLUSTER � � relevance of content PYTHON � � � � � � � � practical relevance � � � � � � CPP � � handout � � � � � � TOTALVIEW � � HPC � � ❯s❡r ❤❛s t♦ . . . the professional competence of the course instructor? � � � � � � LIKWID � � . . . the presentation? VAMPIR � � � � � � � � MATLAB � � ✕ ❡✈❛❧✉❛t❡ t❤❡ ❝♦✉rs❡ s❡ss✐♦♥ . . . the methodical-didactic competence with regard to . . . the structure of learning content ✭❝♦♥t❡♥t✱ t❛❧❦✮ � � � � � � Is there any topic missing and it’s presentation? � � � � � � that you are interested in? ✕ ❝❤♦♦s❡ t❤❡ ❝♦✉rs❡ s❡ss✐♦♥ . . . the participant orientation? � � � � � � . . . the equipment and environment? ✕ ❝❤♦♦s❡ ❢♦❧❧♦✇ ✉♣ s❡ss✐♦♥ � � � � � � . . . the course length? � � � � � � ✭♣②t❤♦♥✱ ❝♣♣✮ . . . the course depth? � � � � � � Follow up courses in Python Do you wish further courses on this subject? � Yes � No ✕ ♣❡rs♦♥❛❧ ❛♥♦♥②♠♦✉s Would you recommend this course? � Yes � No Would you be interested in a Are you using the material later on? � Yes � No follow-up course about . . . ✐♥❢♦r♠❛t✐♦♥ . . . TDD with Python? What did you like about the lecture? � Yes � No ✕ ❝❤♦♦s❡ ②♦✉r ❤✐❣❤ . . . Python project development? ♣❡r❢♦r♠❛♥❝❡ ❝♦♠♣✉t✐♥❣ � Yes � No ❡①♣❡r✐❡♥❝❡ Which ideas and suggestions do you have for this lecture? . . . Other python related topic? Which one? What content would you have additionally preferred? How did you like the exercises? http://csc.uni-frankfurt.de Center for Scientific Computing http://www.hpc-hessen.de Hessisches Kompetenzzentrum für Hochleistungsrechnen

❘❡❢❡r❡♥❝❡ ●✉✐❞❡ Cluster | quick reference | Frankfurt Version 2.0 February 14, 2019 1 Cluster Hardware 3 Software Handling Partitions Cluster Frankfurt Access Cluster Frankfurt run Max Max Max Max Intel Module setting program environments R1 partition time Nodes NodesPU JobsPU SubmitPU Nodes general1 21d 475 150 40 50 skylake ssh <username>@goethe.hhlr-gu.de GOETHE-HLR Syntax : module <command> <modulename> general2 21d 337 150 40 50 broadwell ivybridge avail display all available modules Go to CSC-Website/Access/Goethe-HLR to get an account. The project gpu 21d 50 50 40 50 list display all loaded modules manager has to send a request to submission@csc.uni-frankfurt.de to get CPU-Time for research projects. Further information at our website. test 1h 2-12 10 10 load | add <module> load a module load unstable load a deprecated or unstable module To view such informations on the cluster, use the command: unload | rm <module> unload a module Architecture & Constraints Cluster Frankfurt switch | swap <old-module> <new-module> switch modules sacctmgr list QOS partition format=maxnodes,maxnodesperuser purge unload all currently loaded modules ,maxjobsperuser,maxsubmitjobsperuser # CPUs #nodes CPU GHz RAM Cores scontrol show partition 400 Intel Xeon Skylake Gold 6148 2.10 2/40 192GB sinfo -p partition squeue -p partition 72 Intel Xeon Skylake Gold 6148 2.10 2/40 768GB How-To use custom modules 198 Intel Xeon E5-2670v2 Ivy Bridge 2.50 2/20 128GB 139 Intel Xeon E5-2640v2 Broadwell 2/20 128GB 1 writing a module file in tcl to set environment variables 50 Intel Xeon E5-2650v2 Ivy Bridge 2.60 2/12 128GB 2 module load use.own enables you to load your own modules Per-User Resource Limits Cluster Frankfurt 3 module load ~/privatemodules/modulename Intel Xeon E5-2650v2 Ivy Bridge has 2xAMD FirePro S10000 12GB GPUs. limit description 4 use facilities provided by module MaxNodes max No of nodes The architecture will be selectable via the ’--constraint’ option, MaxNodesPU max No of nodes to use at the same time ivy = dual-socket Intel Ivy Bridge CPU nodes, MaxJobsPU max No of jobs to run simultaneously Installation Spack itself C1 broadwell = dual-socket Intel Broadwell CPU nodes, MaxSubmitPU max No of jobs in running or pending state MaxArraySize max job array size 1001 1 git clone https://github.com/spack/spack.git File Systems storage systems 2 cd spack mountpoint /home /scratch /local /arc[1|2] 2 Cluster Usage Basic Spack Usage size 10GB PU 764 TB 1.4 T 500TB each C2 access time slow fast fast slow How-To-Compile How-To-Run Has to be made after each login system NFS FhGFS ext3 NFS . share/spack/setup-env.sh echo ". spack/share/spack/setup-env.sh" >> ~/.bash_profile network Ethernet InfiniBand Ethernet C1 install spack R1 load module file If you want this to be permanent C2 source spack R2 write bash script - /scratch/<your-project-name>/<your-user-name>/spack/tmp C3 compile software R3 submit job with slurm Contact HPC Frankfurt add to build_stage in config.yaml C4 prepare module file If you have any HPC-questions about SLURM and want help by debugging & optimizing your program, please Getting Help Managing Modules Spack Workflow write to hpc-support@csc.uni-frankfurt.de. Else, you can contact the system administrators if you Cluster Frankfurt Spack https://spack.io/ need software to be installed: support@csc.uni-. . . . You will find further information module avail W1 building packages You will find further information Detailed documentation on using the cluster can be about usable commands on the about usable commands of spack installed sofware packages W2 running binaries found at CSC-Website. clusters with man <command> . with spack --help . spack install lmod W3 developing software to clean things up http://csc.uni-frankfurt.de Center for Scientific Computing http://www.hpc-hessen.de Hessisches Kompetenzzentrum für Hochleistungsrechnen

■t ✐s ♦✉r ❞✉t② t♦ ♦✛❡r ❝♦✉rs❡s ❛t ✉♥✐✈❡rs✐t✐❡s ❛t r❡❣✉❧❛r ✐♥t❡r✈❛❧s ❛♥❞ t♦ ♦✛❡r ✈❛r✐❡t② ✇✐t❤ ♥❡✇ t♦♣✐❝s✳

❈❧✉st❡r ❈♦♠♣✉t✐♥❣ ❈♦✉rs❡ ✶✳ ❈❧✉st❡r ❇❛s✐❝ ✶✳✶ ❈❧✉st❡r ❍❛r❞✇❛r❡ ✶✳✷ ❙♦❢t✇❛r❡ ❍❛♥❞❧✐♥❣ ✷✳ ❙♣❛❝❦ ✷✳✶ ❙♣❛❝❦ ❇❛s✐❝ ✷✳✷ ■♥tr♦❞✉❝✐♥❣ ❙♣❛❝❦ ❈♦♠♣❧❡①✐t② ✷✳✸ ❆❞✈❛♥❝❡❞ ❙♣❛❝❦ ❯s❛❣❡ ✸✳ ❙❧✉r♠ ✸✳✶ ❙❧✉r♠ ❇❛s✐❝ ✸✳✷ Pr❛❝t✐❝❛❧ ❏♦❜ ❙✉❜♠✐ss✐♦♥ ✸✳✸ ❆❞✈❛♥❝❡❞ ❈P❯ ▼❛♥❛❣❡♠❡♥t

❈❧✉st❡r ❈♦♠♣✉t✐♥❣ ❈♦✉rs❡ ✶✳ ❈❧✉st❡r ❍❛r❞✇❛r❡ ✕ ❆❝❝❡ss✴❖r❣❛♥✐③❛t✐♦♥✴❆r❝❤✐t❡❝t✉r❡ t♦ ❛ ❈❧✉st❡r ✕ ❍❛r❞✇❛r❡ ❘❡s♦✉r❝❡s ✭★♥♦❞❡s✱ ❈P❯✱ ★❈♦r❡s✱ ❘❆▼✮ ✕ P❛rt✐t✐♦♥ ✭❙❧✉r♠ ♦♣t✐♦♥s✿ ✲✲♣❛rt✐t✐♦♥ ✱ ✲✲❝♦♥str❛✐♥t ✮ ✕ ❍②♣❡rt❤r❡❛❞✐♥❣ ♦♥ ❛ ❈❧✉st❡r ✭❙❧✉r♠ ♦♣t✐♦♥s✿ ✲✲❡①tr❛✲♥♦❞❡✲✐♥❢♦ ✮ ✕ ❋✐❧❡s②st❡♠ ✕ ●❡tt✐♥❣ ❍❡❧♣ ✭❍❡❧♣✱ ❉♦❝✉♠❡♥t❛t✐♦♥✱ ❙✉♣♣♦rt✱ ❍♦✇✲❚♦✲❈♦♠♣✐❧❡✱ ❍♦✇✲❚♦✲❘✉♥✱ ❍♦✇✲❚♦✲❲♦r❦✢♦✇✮