Data Input/Output in High Performance Scientific Programs

                       Thomas Ludwig
                  Institut für Informatik
                  Universität  Heidelberg
                   t.ludwig@computer.org

Data input/output in high performance scientific programs is a field
of increasing importance. In particular with compute clusters we see
the concept of replicated resources being transferred to I/O
issues. Consequently, we find research questions like e.g. how to map
data structures to files, which resources to actually use, and how to
deal with failures in the environment. The talk will introduce the
problem of parallel I/O from the user´s point of view and illustrate
available programming interfaces. At system level we find parallel
file systems that allow us to efficiently store the data in a set of
individual disks. The talk will give an overview over existing
solutions and current research topics in this field.