Is the bell ringing?

During the Nobel prize-winning UA1 experiment, scientists in the control room used to ring a bell if a particularly interesting event had occurred. Today, the “CMS Exotica hotline” routine produces a daily report that lists the exotic events that were recorded the day before.


Display of an event selected by the Exotica routine.

Take just a very small fraction of the available data (max. 5%); define the events that you want to keep and set the parameters accordingly; run the Exotica routine and only look at the very few images that the system has selected for you. This is the recipe that a small team of CMS researchers has developed to identify the signals coming from possible new physics processes. “This approach does not replace the accurate data analysis on the whole set of data. However, it is a very fast and effective way to focus on just a few events that are potentially very interesting”, explains Maurizio Pierini (CERN), who developed the Exotica hotline routine together with Tulika Bose (Boston University), Massimiliano Chiorboli (University of Catania / INFN / CERN), Leonardo Benucci (Antwerpen University), Elizabeth Twedt (University of Maryland), and Alexey Ferapontov (Brown University).

The amount of data collected every day by the LHC experiments is huge. Despite the very large computing power made available by the Grid, a rough analysis of the whole sample of data produced every day at CMS would take at least a couple of days. “Exotica runs on the so-called ‘express stream’: the small set of data that is used by researchers to control the overall functioning of the detector, that is, about 5% of the total”, says Pierini.

The program sequence starts every day at midnight. The morning after, researchers receive a report listing just the exotic, unusual events. The ten people participating in the project look at the report, which usually contains about 100 events. The data is then read using the CMS Event display that produces a picture to be examined.

But what is the definition of ‘exotic, unusual’ events? “The filter is applied to the energy (high momentum) and the number (high multiplicity) of the particles involved in the events. High-energy phenomena are expected to show new physics. At the beginning, due to the quick increase in luminosity of the LHC, the Exotica routine required fine tuning in order to avoid too large sample of events,” explains Pierini. “Until a few months ago, we were looking for typical Standard Model events like Z and W boson decays. Now those events have become quite usual and Exotica is fixed to hunt out completely new phenomena. For example, during the first phase the report gave warning of events with 2 muons and energy over 15 GeV. Now the lower limit is set at 50 GeV”. At this level, we should see the Higgs boson!

The Exotica procedure reverses the traditional method where the final images are produced after an in-depth analysis has been carried out on the whole set of data. In this approach, statistics define the ‘interest’ of a particular event. “Exotica – on the other hand – does not produce statistics, it is just a numerical filter that spots exceptional events, and it is an early warning device for any complex and rare effects or detector malfunctioning. It has already been used to fix several subtle problems in the event reconstruction. Exotica cannot substitute analysis but it is a useful tool that allows us to come across unexpected events. In such cases, an in-depth analysis will have to be carried out to fully understand the process”, concludes Maurizio Pierini.

by Francesco Poppi