Bonn 2010 – wissenschaftliches Programm
HK 10.9: Vortrag
Montag, 15. März 2010, 18:30–18:45, HG ÜR 6
FPGA-based Adaptive Computing for Online Trigger Algorithms — •Ming Liu1, Axel Jantsch2, Andreas Kopp1, Wolfgang Kühn1, Jens Sören Lange1, Yutie Liang1, Zhonghai Lu2, David Münchow1, Björn Spruck1, and Qiang Wang1 — 1II. Physikalisches Institut, Universität Gießen — 2Dept. Of Electronic Systems, Royal Institute of Technology (KTH), Sweden
Conventionally, FPGA-based static implementation of trigger algorithms results in shortcomings including low resource utilization efficiency, management complexity of design modules in large projects, etc. By contrast, online adaptive systems can intelligently schedule and manage various algorithm modules during run-time. Adaptive computing not only enables the integration of multiple algorithms in small FPGA chips to reduce the hardware cost, but also achieves higher performance-price ratio compared to the static design approach. It also simplifies the design management and maintenance work during experiments. In our work, we propose an adaptive design framework based on the FPGA Partial Reconfiguration (PR) technology. Several key aspects have been respectively addressed in hardware/software layers. The adaptive framework is foreseen to be general-purpose and beneficial to the data acquisition (DAQ) and trigger systems in different experiments such as PANDA, HADES, WASA, etc.
This work was supported in part by BMBF under contract Nos. 06GI9107I and 06GI9108I, FZ-Juelich under contract No. COSY-099 41821475, HIC for FAIR, and WTZ: CHN 06/20.