Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

T: Fachverband Teilchenphysik

T 76: Grid-Computing I

T 76.6: Vortrag

Mittwoch, 11. März 2015, 18:00–18:15, K.11.20 (K5)

Improvements for PanDA and intermediate layers — Gen Kawamura, Erekle Magradze, Haykuhi Musheghyan, Jordi Nadal, Arnulf Quadt, and •Gerhard Rzehorz — II. Physikalisches Institut, Georg-August-Universität Göttingen

The PanDA Production and Distributed Analysis system is handling the ATLAS workload management for production and distributed analysis processing. It was designed for analysis as well as production for High Energy Physics. One of the advantages of this pilot based system is, that it has its own integrated monitoring solution. Monitoring is a method used in computing, it means that a certain process is observed and overseen and is usually also protocolled. In the case of a process or system failure, a responsible person should be notified and countermeasures taken. Since PanDA is very versatile, it can also be used to process jobs on a computing Cloud, instead of just using the Grid(WLCG). Cloud computing resources can be provided by private companies, that bill the resources that are actually being used, for example CPU power over time. The advantage of this is obvious if one looks at the cost. They are basically the same, whether a huge task is done on little CPU power over a long time or on plenty of CPU power in a short time. For these reasons, a close monitoring, for example by the PanDA system for the usage of Cloud resources is important.