About
A novel architecture is being proposed for the data acquisition and trigger system for PANDA experiment at the HESR facility at FAIR/GSI. The experiment will run without the hardware trigger signal and use timestamps to correlate detector data from a given time window. The broad physics program in combination with high rate of 2 10^7 interactions require very selective filtering algorithms which access information from almost all detectors. Therefore the effective filtering will happen later than it used to in today's systems ie after the event building. To assess that, the complete architecture will be built of two stages: the data concentrator stage providing event building and the rate reduction stage. For the former stage, which allows to switch 100 GB/s of event fragments to perform event building, we propose two layers of ATCA crates filled with compute nodes - modules designed at University of Giessen for trigger and data acquisition systems. Each board is equipped with 5 Virtex4 FX60 FPGAs and high bandwidth connectivity is provided by four Gbit Ethernet links and 8 additional optical links connected to RocketIO ports.
Using the SystemC as the modeling platform we designed simplified models of the components of the architecture and demonstrated expected throughput. We also show impact of some architectural choices and key parameters on the architecture's performance..
0:00:00 Slide 1
0:01:16 Slide 2
0:02:31 Slide 3
0:05:01 Slide 4
0:07:31 Slide 5
0:10:01 Slide 6
0:12:32 Slide 7
0:15:02 Slide 8
0:17:32 Slide 9
0:20:02 Slide 10
The talk
Credits and licence
- Institution
- CERN
- Department
- Conferences
- Speaker
- Krzysztof Korcyl
- Year
- 2012
- Language
- English
- Subjects
- Physics, Computer science, Earth and environment
- Source
- CERN
- Licence
- CERN, as CERN states it. CERN's terms
- Attribution
- © CERN. CERN provides its audiovisual media free of charge for educational and informational use, credited to CERN; this lecture is shown in CERN's own player.
- Who can watch
- CERN gives no rating. Invisible Network shows it to Tweens profiles and older, judged from what the source says about it. How courses are placed