Banner
    Recent Results Of CMS
    By Tommaso Dorigo | June 14th 2012 01:28 AM | 10 comments | Print | E-mail | Track Comments
    About Tommaso

    I am an experimental particle physicist working with the CMS experiment at CERN. In my spare time I play chess, abuse the piano, and aim my dobson...

    View Tommaso's Profile
    Two days ago I discussed at ICFP 2012 the most recent results of the CMS experiment at the CERN Large Hadron Collider. In the allotted time of my talk I could only cover few analyses, and I obviously chose some of the most interesting ones, so that was already a summary. Here I am bringing the information collapse one step further, by giving a itemized summary of some of the points I made, just in case you are interested. If you want to, you can also download the original slides of my talk from here (but be careful, it's a 8Mb file).

    - The LHC has yielded over 5 inverse femtobarns of proton-proton collisions to CMS to analyze in 2011, and these data have been used for dozens of new results. Now we have on tape another 5/fb of data from the 2012 run, but these have not been looked at yet (results will be ready in a few days).

    - We can broadly divide 2011 results into three areas: searches for Higgs bosons, Standard Model measurements, and new physics searches.

    - CMS searched for the Higgs boson in eight independent final states, further divided in over forty categories. The combined results of these searches say that the particle must be lighter than 127 GeV (and heavier than 115 according to LEP II), or heavier than 600 GeV. We know the latter is not an option as far as the Standard Model is concerned, because it would now be utterly inconsistent with other electroweak measurements. So we might argue that if the SM Higgs exists, we already know its mass to better than 10% accuracy.

    - CMS finds a signal with a local significance of 3.1 standard deviations at 124 GeV. If this excess is due to the Higgs boson, it is likely that the new data, once analyzed, will produce additional evidence which can be considered conclusive proof for the particle's existence.

    - A new baryon, the Ξb*, has been observed in its fully exclusive cascade decay into J/ψ, proton, and pions (with intermediate Ξb and Λ states). Its mass is just short of 6 GeV (see picture on the right, showing the peak in the distribution of Q-value of the two-body decay Ξb*-->Ξb π).
     
    - Rare decays of the Bs meson have been searched, and a tight limit on the Bs->μμ decay has been obtained by combining CMS results with LHCb and ATLAS ones. New physics models are strongly constrained by this limit because many realizations of NP would yield enhancements in the branching ratio for the dimuon decay mode.

    - CMS now measures the top quark mass and cross section in a number of different techniques. The precision on the top mass is reaching the Tevatron average (1.25 GeV total error now). A new era of precision top physics measurements has started, with e.g. limits on Flavour-changing neutral current top decays constrained at the 0.34% level, and top-antitop mass difference measured to within 0.5 GeV (of course it is zero!).

    - A large number of interesting searches for new physics returned null results. Supersymmetry has been investigates in dozens of possible signatures, with no positive result.

    Below is my conclusions slide:

    Comments

    Would it be possible for you to provide the slides in pdf format?

    /Thanks a bunch from a reader not using Power Point

    dorigo
    Hi,
    okay, here they are: http://www.pd.infn.it/%7Edorigo/icfp2012.pdf
    4Mb file.
    Cheers,
    T.
    BDOA
    Cool. Opposite to a flavour changing neutral current, hints of W mediate decays that are not equivalent for all output leptons have been seen in Bs-> mu + mu bar, can we look for this in Xi*b decays, which also have the strange and bottom quarks. As far a I know everywhere else the generation symmetry holds for quark and for leptons separately, but I have not seen published limits for this.
    BDOA Adams, Axitronics
    Tommaso, your slide 38 of 43 (Search for V+jj Bumps) says:
    "... Two years ago CDF published a greater-than 4 sigma evidence of a structure ...
    CMS looked for this dijet resonance with 2011 data ... No signal is observed ...
    95 percent CL upper limits exclude the CDF effect ... see arxiv:1107.4771 ...".

    However, the middle plot at the bottom of your slide
    a copy of which I put up at tony5m17h.net/WjjCDF2011CMS.png
    shows
    that CMS has a data point clearly above background plus uncertainty
    which data point is where a "CDF-like Signal" would be expected.

    Is that consistent with a CDF-like V+jj phenomenon with a cross section somewhat lower than than the 4 pb stated in the CDF paper at arxiv 1104.0699 as "the order of" the cross section observed by CDF ?

    To be more specific,
    are the CMS results consistent with a CDF-likeV+jj phenomenon with cross section a bit less than 3 pb ?

    Tony

    How long will it take for the 5/fb collected now to be analyzed for Higgs signals?

    dorigo
    A day.
    I will be more specific soon...

    Cheers,
    T.
    One day? Does it mean that you will collect data up to July 6, for the July 7 conference?

    dorigo
    No, rather that spinning the new data does not take very long. What takes longer is to validate all results, combine them, and approve them. But the reconstruction of the data takes a while longer, so in fact right now only a part of the data already collected is available for analysis.

    Cheers,
    T.
    Tommaso,

    can you explain what the difference between "spinning" (same as analyzing?) and "reconstructing" (a while longer) and "analyzing" (one day) is?

    Erik

    dorigo
    Hi Erik,

    "spinning" (aka analyzing)  is when we take our analysis program and we get it to read all the (reconstructed) event files, producing final histograms and results. Reconstructing comes earlier, when raw data (from detector components, e.g. silicon hits and calorimeter energy deposits) is used to reconstruct physics "objects" like tracks of charged particles, muon and electron candidates, jets, photons, tau candidates, missing transverse energy, primary vertices. This phase takes much longer. You could very roughly say that one CPU takes O(1s) or even longer to reconstruct an event, while a typical analysis job will take O(1ms) to do even elaborate things with the event data.
    Now take 5/fb of data and ask yourself what this means in terms of events: it is not straightforward to compute how many events that is -it would require knowing the exact cross section written to tape for all the different "data streams" (corresponding to the trigger having accepted the event based on the presence of different candidate objects, e.g. muons or electrons, high-pT jets, etcetera). But very roughly, if 5/fb were taken in three months of data taking, with a duty cycle of 30% that is 3 million seconds of data written to tape at 300 Hz rate -> a billion events.
    So you could say that one CPU takes O(1Gs) to reconstruct the data; a thousand CPUs will take two weeks to do that. These are order of magnitude estimates, so take them with a pinch of salt. Also, I am not taking into account the intermediate steps (calibration procedures need to be run to get the corrected energy estimates, for instance, and these require running calibration datasets first... But let's omit these details).

    Now for the spinning: a Higgs to gamma-gamma search will only look at a subset of those billion events -those recognized to contain photon candidates. Say these are 50 million events: it will take half a day to produce results by "spinning" them, if you program takes 1 ms for each event, and if you run a single CPU. Again, these are very rough estimates and only show the "typical" time expense. For the Higgs to gamma gamma in fact the analysis takes longer because there are some neural networks performing complicated energy corrections... Every analysis has its own bottleneck. But the orders of magnitude are the ones I described above.

    Note that the LHC experiments are much more effective in the reconstruction of data, both thanks to improved detectors and algorithms and to larger CPU pools. In CDF I remember that the delay between taking a data event and having it available after reconstruction was of the order of four to six months; at the LHC this is much less painful.

    Cheers,
    T.