Parameter studies on heterogeneous computing infrastructures with the Scalarm platform
AUTHORS: D. Król, R. Słota, J. Kitowski
ABSTRACT
The main goal of this tutorial is to demonstrate the Scalarm platform as a tool supporting parameter studies on different computing infrastructures like clusters, grids and clouds. Parameter study (also called parameter sweep) is an approach where the same application is executed many times with different input parameter values. Afterwards, results from every execution are collected to identify trends or anomalies in the data that may lead to new insights. Despite their simplicity, parameter studies are commonly used in many fields of science, including computational fluid dynamics, particle physics, and discrete-event simulation. It is arguable the most popular way of scientific computing organization.
