At long last, the LHC today ran a rather interesting test of “high pileup” conditions. A quick reminder about pileup: the beam at the LHC (and all particle-physics accelerators) is bunched rather than continuous. Each time a bunch in one beam passes by a bunch in the other, multiple protons can interact with each other. It’s rare for more than one of these interactions to be a “hard scatter” and thus be likely to produce interesting particles. The other interactions still happen, though, and you’ll have some number of new particles produced from each of those. That’s what we refer to as the pileup. (I never liked the term, as “pileup” is also something that can happen in the electronics that we use for the detectors and the overloaded definition is confusing. I think “multiple interactions” describes it much better.) Of course, remember that our ultimate goal is to maximize the number of interesting collisions, and one way to do that is to maximize the number of particles per bunch and thus the number of interactions per crossing…but that that means more pileup, too.
As the LHC has been running lately, the typical number of interactions per crossing at the start of a fill (when the beam intensities are highest) is about 15. But in 2012, the LHC will probably run with even more particles per bunch, resulting in more interactions. Can the detectors, triggers and software handle these bigger and busier events? Today we had a test; the LHC ran with fewer bunches than usual, but with many more protons per bunch. Here is the number of interactions per beam crossing as measured by the CMS online monitoring system:
Pileup vs. time for today's high-pileup fill
As can be seen, the fill started at 32 interactions per crossing — about twice as much as we have during regular LHC running now. (There were a couple of intervals in which the LHC separated the beams at CMS, as a test, and the pileup number drops then.) At the end of the (short) fill, we still had 25. We’ll be using this data to help understand how to best optimize our operations for next year. Next year isn’t far away — the proton-proton run ends this weekend!
