Creation
of global e-Infrastructure involves an integration
of isolated local resources into common heterogeneous
computing environment. In 2014 a pioneering work to develop a large scale data- and task- management system for federated heterogeneous resources has been started at the National Research Centre “Kurchatov
Institute” (NRC KI, Moscow). As a part
of this work, we have designed, developed and deployed a portal to submit payloads to heterogeneous
computing infrastructure. It combines Tier-1, Cloud-infrastructure, and a supercomputer at the Kurchatov
institute. This portal is aimed to provide a common interface to submit tasks to Grid sites, commercial and academic clouds and supercomputers. Integration
of Tier-1 and the supercomputer has allowed to notably increase total CPU capacity available for Large Hadron Collider (LHC) experiments. The portal can be used not only for High Energy Physics (HEP) applications, but also for other compute-intensive sciences such as bioinformatics with genome and sequence analysis; astrophysics with cosmic rays analysis, antimatter and dark matter search, etc.
This article describes developed portal as a top layer for computing facilities infrastructure for High Energy Physics and other compute-intensive science applications. The article presents the results of using PanDA at NRC KI supercomputer/Cloud as underlying technology for the portal.