Howdy! LHC fans, and welcome to my US LHC blog area. It is a great pleasure for me to have this opportunity to communicate what we are up to at the LHC from a more personal perspective. I hope you enjoy it!!
This is my first post and so I thought it was a good idea to talk about things related to my first HLT (High-level Trigger) expert shift. It was a few weeks ago when the HLT cell phone rang in my pocket for the first time with a phone-call from the CMS control room. My plan of attending a French lesson in Geneva later in the evening suddenly changed for a more exciting one. There was a problem with the HLT system and, as the expert on-call, I needed to figure out the nature of the problem very quickly. Before telling you about the problem, let me try to explain a key feature of our trigger system: regional reconstruction.
Imagine a friend of your identical twin brothers hands you over two pictures of his group of friends. One of them is in front of the Pantheon in Rome and the other one in front of the façade of the Pantheón in Paris. Now, you have never been in either city and you know almost nothing about these historical sites (they look almost the same to you), but he challenges you to identify the cities in one minute. Since you are clever and you know that only one of your twin brothers was able to make it to these two cities at a time, you rapidly identify them and correctly associate the cities. Your friend is very impressed!
Now, imagine the same friend hands you over the same two pictures but made into finely-cut jigsaw puzzles and he challenges you again with the same task. You have only one minute to identify the cities but now the faces of your brothers may not look as identifiable as they were before. However, you remain clever and take the whole minute to try to reconstruct the face of who you think is one of them in one of the photos. You start by a “seed” (a puzzle piece containing your brother’s eye, for instance) and then you try to arrange the pieces around the seed to quickly take a look at his face. The task was not easy because there were many pieces, but you were fast enough to complete one face in one picture and, therefore, identify the cities correctly once again. Your friend is shocked by how smart you are!!
The electronic signals of the millions of independent channels in a modern particle detector are like jigsaw puzzle pieces of a picture of a collision. As there will be billions of them happening every second at the LHC, we use a system called trigger in order to select only the interesting collision “photos”. In CMS, once first beam collisions arrive, we will not have enough time to look at the whole collision photo, hence, similarly to the analogy above, we put together just a couple of “particle faces” (regional reconstruction) using “puzzle pieces” from one or a few sub-detectors in order to be able to say (or not): “Yes, I recognize this face (I have seen it before), I will be interested in this picture, let’s keep it; we can put all the pieces back together later to see who else was in it and what the background was like (full reconstruction)”.
The problem I had to handle, after receiving the phone call, had to do with the reconstruction of some of the “faces” in our pictures of cosmic rays data taking a very long time and clogging our system. What makes it very exciting is that, in case of a problem with any subsystem in the detector, an on-call expert needs to react very fast, usually under a lot of pressure. This is because if we stop taking “photos” for any reason, we might miss the “kodak” moment of a lifetime: a black hole, a Higgs boson, a supersymmetric particle, or any other exotic event. We use cosmic data to better understand our system and to prepare for beam collisions.
We will be ready at CMS for our ultrafast photography adventure!
Edgar Carrera, Boston University
