This year course is over. Thanks everyone who has made it succesfull! Keep watching SCiP pages, more courses are coming.

Scientific Computing in Practice: kickstart 2014 course

Part of Scientific Computing in Practice lecture series at Aalto University.

Audience: Aalto employees and students looking for extensive intro into scientific computing facilities at Aalto.

Course purpose: kickstart for participants to get started with the available computational resources at Aalto. Inroduce CSC and FGI HPC resources. Give hints, ready solutions and copy/paste ready examples on how to compile and run available software and how to develop your own code. Leaving the course you expect to know where you suppose to run your job, as well as how to use available software or install your own, and how to make it in the most efficient way on Aalto resources. In addition to general introduction, this year kickstart will include GPU programming and Python for Scientific Computing parts.

Time, date:  Mon 9.6 — Wed 11.6.2013, from 9:00 – 17:00 all days

Place: Otaniemi, Otakaari 1, D-hall (main building, main entrance, first floor)

Days’ time schedule is flexible, here is the tentative plan. 45 minutes break around 12:15-13:00, plus short 10 minutes breaks in between.

Day one:

  • HPC Fundamentals: terminology, architechtures, interconnects, infrastructure behind, world-live examples
  • Linux cluster at School of Science: Triton practicalities
    • environment: module, available space
    • SLURM: effective use of batch system
    • Lustre: how to make your I/O work smoothly
    • MATLAB: large memory, GPU and other runs with Matlab on Triton, live demos
    • serial runs: how to run your serial jobs
    • parallel runs: what is behind the MPI libs, and how do not get confused with OpenMPI vs OpenMP

45 minutes break

  • Compiling/debugging your code: where to find optimized libs and prepare your Makefile, how to benchmark and debug the code. Profiling tools available on Triton mentioned.
  • Grid computing: FGI resources available for free (by Kimmo Mattila, CSC)
  • CSC resources intro: most powerful computational resources in Finland available to researchers (by Tomasz Malkiewicz, CSC)

Day two:

full day, break in between

  • Python for scientific computing (by Jussi Enkovaara, CSC)
    • Numerical calculations with NumPy
    • Visualization with matloblib
    • Overview of scientific tools in SciPy
    • Parallel programming with Python and mpi4py

Day three:

full day, break in between

  • GPU computing (by Topi Siro, Aalto)
    • Running GPU programs on Triton
    • Basics of CUDA and OpenACC
    • Optimizing CUDA codes

Lectures: Janne Blomqvist (D. Sc.), Mikko Hakala (D. Sc.), Ivan Degtyarenko (D. Sc.), Topi Siro (Doctoral Candidate)

Invited Lectures: Jussi Enkovaara, CSC (D. Sc.), Kimmo Mattila, CSC, Tomasz Malkiewicz, CSC (D. Sc.)

Registration: free of charge, registration is required, we want to make sure we have enough space for all participants.

Cost: Free of charge for Aalto employees and students

Credits (opintopisteet): Course certificate available on request. Full course hours correspond roughly to 1 ECTS.

Other comments: All parts will include live examples and demos and tutotials (!). Participants will be provided with access to Triton for running examples. Participants are expected to bring their laptops with the pre-installed SSH client, configured for using WiFi (preferably Eduroam, otherwise Aalto Open).

Additional course info at: Ivan Degtyarenko, ivan.degtyarenko at aalto.fi

Course materials:

Exersices examples and answers are on Triton at /triton/scip. Check out kickstart/, python/ and gpu/ subdirectories.