Before telling you what I am doing now I would like to spend a few words on what I did before being ALICE electronics co-ordinator. The first years of my CERN life I spent designing the fast CMS muon trigger track finder processor, which looked for those four muons in the CMS detector with the highest transverse momentum within 350 ns and of course for each of the 40 millions of bunch crossings per second. My trip to fast electronics was continued when I joined briefly the ALICE time-of-flight (TOF) collaboration to take part in the development of sub-100 ps resolution TOF detectors before I settled for quite some time with the ALICE silicon pixel detector project (SPD). There I started developing ASICs for the on detector read-out of the detector and ended up coordinating the design, development and installation of the SPD read-out and trigger system. Since the SPD electronic system seemed to have worked well between the operation startup of ALICE in 2008 and the start of upgrade efforts I continued my fast electronics lane and coordinated and even took part (yes this still exists, coordinators who take part in detailed work) in the research and development of a pixel detector with less than 200 ps time resolution for the GigaTracker detector in the NA62 experiment. This allowed me to combine my interest in fast time tagging detectors (TOF), pixel detectors (SPD) and fast read-out processors (CMS muon trigger) leading to a prototype ASIC with 170 ps time resolution and a full scale read-out ASIC which still needs to be tested.
This brings me right away to today. I was given the opportunity and fantastic challenge to be part of the upgrade of our ALICE experiment as electronics co-ordinator. Thus, as you can guess, my road on the fast electronics lane is not finished yet, because upgrading ALICE means we are going to read out and process huge amounts of data, even more than now, in very fast time. On top of this we need to implement the system very quickly as the installation of the upgraded ALICE detector is foreseen for 2018. My task is to find the way through the maze of technical and organisatorial options to read out and trigger our detector which is suited best for our collaboration in terms of performance and resources.
Compared to today ALICE will see more than 5 times as many Pb-Pb collisions (from 8 kHz to 50 kHz) and on top of that we will read out and store ALL of the collision data for off-line analysis. We will still have a central trigger processor being the master brain of the ALICE read-out, however, the trigger or event selections algorithms will be optimised for read-out efficiency and less for physics event selection. We are going to upgrade most of the ALICE systems. We will have the first continuous read out TPC, requiring no trigger signal at all and sending the data in a continuous stream, like in a movie, to the on-line computers. The other detectors will be triggered, some of them being completely new exciting developments, such as monolithic pixel detectors (ITS, MFT), others where the chambers stay in place but the electronics will be upgraded. Our new system architecture integrates the triggered and non-triggered data streams to be analysed later on. As in ALICE there are many, enthusiastic experts to develop systems, but even these experts are available only in finite numbers, we are continuing the successful ALICE approach to develop common items. It used to be the detector detector link system (DDL) and this approach will be widened to include the off-detector read-out units (common read-out unit - CRU) and the new development of the read-out ASIC (SAMPA) will be shared by TPC and the muon chambers. No doubt, the central trigger processor (CTP) will be upgraded and again provide a central system to create and distribute the timing and trigger signals.
At the end, what is left for us to do, is to put the common and dedicated developments together, so that, when the long shut-down 2 (LS2) is over, we switch on the system, enjoy and benefit from its performance.