Today, on the third day of the WikiToLearn Sprint at CERN hosted by KDE e.V., we had the pleasure of listening to an interesting and inspiring lecture by Professor Pere Mato Villa, who talked about Computing for Data Processing and Analysis at CERN. In approximately one hour, we were enlightened on the techniques and methods in use in the various LHC experiments to acquire and process raw data from detectors. He also explained the massive extent of the IT infrastructure that’s needed to host all the data: currently all the LHC experiments rely on distributed computing resources, accounting for roughly 350,000 CPU cores, and 400 PB of disk and tape storage combined. That’s a huge one!
Together with the hardware goes the software: in 1995 the ROOT project started and it has since then helped many scientists in speeding up their workflow. Another cool software that he discussed was Geant4, a toolkit for the passage of particles through matter.
Eventually, he prospected the new challenges that LHC experiments will face: mainly going Big Science on the Cloud, since cloud resources will follow the evolution of bare-metal ones, and adjusting the software to these changes.
I am personally astonished as I learn more and more about the great efforts and the fantastic results that are achieved here at CERN, and I sincerely hope that we’ll be looking in future at more and more people willing to make a difference for science.