Knowledge Base: Tutorials and Manuals

Manuals Intel (GPC): C & C++ compiler Fortran compiler Intel Compiler Floating Point Consistency Math Kernel Library (MKL) Math Kernel Library's Vector Statistical Library Intel MPI library Math Kernel Library link line advisor ($MKLPATH → ${MKLPATH} in makefiles) 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 (with error codes) LoadLeveler: using & administering Scheduler (Adaptive Computing/Cluster Resources) Moab workload manager Moab commands Torque resource manager Torque PBS commands

Training material

For upcoming classes, see our Courses website!

SciNet Basics

• Essentials
• SciNet User Tutorial
• Intro to SciNet, SciNet, Sept 2010
• Frequenty asked questions
• Ssh
• GPC quickstart guide
• TCS quickstart guide
• Important .bashrc guidelines
• Software and libraries modules system

Batch job management

• Wallclock time
• Checkpoints
• Signals
• Moab
• Serial Jobs
• Ramdisk
• Moab workload manager
• Moab commands
• Torque resource manager
• Torque PBS commands
• Parallel environment
• Cluster information center (with error codes)
• LoadLeveler: using & administering

Programming

• Intel C & C++ compiler
• Intel Fortran compiler
• IBM AIX Workshop, SciNet, Nov 2008
• IBM MPI Workshop, SciNet, Feb 2009
• IBM Compiler Workshop, SciNet, Feb 2009
• IBM C compiler, language, optimization
• IBM C++ compiler, language, optimization
• IBM Fortran compiler language, optimization
• IBM UPC compiler, language, optimization, library, user's guide, programmer's guide
• Perl
• Oct 2010 SNUG TechTalk: Version Control
• Software Carpentry Resources

Parallel Programming

• One-day Intro to Practical Parallel Programming, SciNet, 22 Sept 2010:
  • Morning Slides, Intro and OpenMP
  • Afternoon Slides, MPI
  • Example source code
• Five-day Parallel Scientific Computing Workshop, SciNet, Aug 2009:
  • Slides
  • Video
• 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
• LLNL MPI Tutorial: This was the basis for the MPI workshop at SciNet.
• Intel MPI library
• GPC MPI Versions
• Co-array Fortran on the GPC

Performance Tuning

• Introduction To Performance
• Performance tools for GPC and TCS
• Dec 2010 SNUG TechTalk: Profiling Tools on GPC
• 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.
• Performance tuning

Debugging

• TechTalk: Debuggers & Parallel Debugging on SciNet - gdb, ddd, padb, SciNet User Group Meeting, Nov 2010

Math libraries (BLAS, LAPACK, FFT)

• Intel Compiler Floating Point Consistency
• Math Kernel Library (MKL)
• Math Kernel Library's Vector Statistical Library
• Math Kernel Library link line advisor
  ($MKLPATH → ${MKLPATH} in makefiles)
• ESSL high performance math library V4 (V3)
• Parallel ESSL high performance math library V3.3

I/O

• Data management
• Intro to Parallel I/O, SciNet, Oct 6th, 2010:
  • Morning & MPI-IO Slides
  • NetCDF/HDF5 Slides
  • Source Code.
• Sept 2010 SNUG TechTalk: Parallel File System and IO
• File System and I/O dos and don'ts

GPU Computing

• SciNet GPU Workshop July 2010
• NVidia archived courses for GPGPU Programming

Visualization

• Using Paraview

Applications

Currently, information on the following applications is available:

• Athena
• Amber
• CCSM3/CCSM4
• FLASH3
• Gamess. See also here
• Gromacs
• LAMMPS
• NAMD
• Paraview
• VASP
• See also User Codes