Bereiche | Tage | Auswahl | Suche | Downloads | Hilfe

HK: Fachverband Physik der Hadronen und Kerne

HK 36: Postersitzung

HK 36.76: Poster

Mittwoch, 17. März 2010, 14:00–16:00, HG Aula

Control and Management Unit for a Computation Platform at the PANDA Experiment^* — Martin Galuska, •Thomas Geßler, Wolfgang Kühn, Johannes Lang, Jens Sören Lange, Yutie Liang, Ming Liu, Björn Spruck, and Quiang Wang — II. Physikalisches Institut, Justus-Liebig-Universität Gießen

The FAIR facility will provide high intensity antiproton and heavy ion beams for the PANDA and HADES experiments, leading to very high reaction rates. PANDA is expected to run at 10-20 MHz with a raw data output rate of up to 200 GB/s. A sophisticated data acquisition system is needed in order to select physically relevant events online.

For this purpose a network of interconnected compute nodes can be used. Each compute node can be progammed to run various algorithms, such as online particle track recognition for high level triggering. An ATCA communication shelf provides power, cooling and high-speed interconnections to up to 14 nodes. A single shelf manager supervises and regulates the power distribution and temperature inside the shelf.

The shelf manager relies on a local control chip on each node to relay sensor read-outs, provide hardware adresses and power requirements etc. An IPM controller based on an Atmel microcontroller was designed for this purpose, and a prototype was produced. The neccessary software is being developed to allow local communication with the components of the compute node and remote communication with the shelf manager conform to the ATCA specification.

^*This work was supported in part by the BMBF (06GI9107I and 06GI9108I) and HIC for FAIR.