Computational Modelling Group

Workshop  26th September 2012 9 a.m.   Culham Centre for Fusion Energy, Oxfordshire

Parallel I/O Course

Web page
http://www.hector.ac.uk/cse/training/parallel_io/
Categories
C, C++, Distributed computing, e-Research, Fortran, HPC, MPI, Optimisation, Scientific Computing
Submitter
Jess Jones

NAG HeCTOR Training

Description: Input and Output (I/O) is often an under-considered part of a code but can severely limit its scalability, particularly where data is collected on to one process before being written to disk. This course will present MPI-IO, introduced in the MPI 2.0 standard, which allows a program to read and write to a single file from multiple processes. We will also take a look at the parallel I/O libraries NetCDF and HDF5, which are built on top of MPI-IO. The course will include practical exercises and a case study.

Aimed at: Experienced MPI programmers who would like to learn how to do their I/O in parallel.

Prerequisites: Attendees should be able to program in either Fortran or C and be familiar with working in a UNIX environment (i.e., you should be able to connect to a machine remotely, use basic UNIX commands, edit a source file and understand the elementary steps in compiling object files and creating executables). Attendees should also be familiar with MPI, including the construction of derived datatypes and ideally, communicators.