I

Through this public-private partnership, the researchers prepared the techniques necessary for acquiring and processing hundreds of terabits per second using and expanding the most innovative concepts available in the ICT industry today. They explored new areas, including silicon photonics for network links in harsh operational conditions, and tight integration of reconfigurable logic with commodity processors to bring new approaches to data acquisition - all with increased performance and decreased cost in mind.

“The ICE-DIP project is an excellent example of how research can collaborate with industry to train the scientists, engineers, and ICT specialists of tomorrow,” said Eckhard Elsen, director for research and computing at CERN. “Education is a core part of CERN’s mission. As well as the wider societal benefits that such training brings, we’re always keen to absorb and integrate bright ideas and fresh new perspectives in tackling the challenges of our evolving research programme.”

Here are summary presentations from each of the students, and links to their theses or published papers.

• Grzegorz Jereczek,, Software Switching for High Throughput Data Acquisition Networks

• Marcel Zeiler, Radiation-hard Silicon Photonics for Future High Energy Physics Experiments

• Srikanth Sridharan, FPGAs for next gen DAQ and Computing systems at CERN

• Aram Santogidis, Data transfer on manycore processors for high throughput applications

• Przemyslaw Karpinski, The consequences of poor vectorization