Knowledge Base: Tutorials and Manuals

Tutorials and workshops SciNet Class Schedule Let us know which one(s) you will attend by email All classes will be given at the SciNet headquarters. "Intro to SciNet" classes. These will take about 90 minutes of your time and you will learn a lot on how to use SciNet's systems. Even experienced SciNet users may still pick up some pointers. - Fri 3 Sept, 12-1:30. First of the year, with Pizza! - Fri 10 Sept, 12-1:30. Please bring a print-out of the latest User Tutorial to the class. More will be given throughout the year depending on user interest. TechTalk: These half-hour sessions will be part of the SciNet User Group meetings, every second Wednesday of the month at 12 noon. SNUG meeting will also include user discussion and pizza. TechTalk Topics: - Wed 8 Sept: The GPFS File Systems - Wed 13 Oct: Version control on SciNet - svn, git, mercurial - Wed 10 Nov: Debuggers & parallel debugging on SciNet - gdb, ddd, padb - Wed 8 Dec: Performance and profiling on SciNet - gprof, scalasca, peekperf Classes - Sept 22: Intro to Parallel Programming (whole day) - Oct 6: Parallel I/O (whole day) - May 2011: 5-day Parallel Programming course. OpenMP, MPI and more will be covered. The aim is for you to leave the course able to start parallelizing your code. SciNet User Tutorial SciNet GPU Workshop July 2010 Parallel Scientific Computing Workshop, SciNet, Aug 2009 Slides /Video IBM AIX Workshop, SciNet, Nov 2008 IBM MPI Workshop, SciNet, Feb 2009 IBM Compiler Workshop, SciNet, Feb 2009 External: Virtual School for CSE Web courses (Jul/Aug 2010): Petascale programming environments and tools Big data for science Proven algorithmic techniques for many-core processors NVidia archived courses for GPGPU Programming LLNL MPI Tutorial This was the basis for the MPI workshop at SciNet. High Performance Computing Book Online version of an older O'Reilly book which covers the basics of (mostly serial) programming for performance. Covers the most important issues today very clearly. How to Write Fast Numerical Code Good introduction to thinking about performance.

Manuals Intel (GPC): C & C++ compiler Fortran compiler Intel Compiler Floating Point Consistency Math Kernel Library Math Kernel Library link line advisor Intel MPI library IBM (TCS): C compiler, language, optimization C++ compiler, language, optimization Fortran compiler language, optimization UPC compiler, language, optimization, library, user's guide, programmer's guide ESSL high performance math library V4 (V3) Performance tuning Parallel environment Cluster information center (also explains error codes!) Scheduler Moab workload manager(Adaptive Computing/Cluster Resources) Moab commands (Adaptive Computing/Cluster Resources) Torque resource manager (Adaptive Computing/Cluster Resources) Torque PBS commands (Adaptive Computing/Cluster Resources) LoadLeveler: using & administering (IBM)