Pictures from the Control Room

by

When I was last at CERN (last month), I took some pictures of the CMS Control Room, to give you a sense of the place.  Some control rooms (also called “counting rooms” for historical reasons) show their age and cumulative history, like the CESR control room at Cornell, which has oscilloscope displays next to flat screens, a built-in AM car radio, and an old Macintosh to read all the alarm messages over the intercom in a robotic voice.  I’ve never seen one yet that looks like something from a James Bond movie, but the “CMS Centre” comes close.

The CMS Centre is not CMS’s only control room, and in fact it’s not the one which is most often called The Control Room.  CMS is located on the opposite side of the ring from the main CERN site, a 20 minute drive away, and the Control Room proper is located there, 100 meters underground, in a cavern adjacent to the experiment.  To get to it, one must pass through a gate controlled by retinal scan, down an elevator, and through downward-sloping, white-painted tunnels, wearing a hard hat and steel-toed boots.  Okay, maybe that’s more James Bondy.  But in the Control Room, it’s mostly the low-level aspects of the detector that are monitored: if a board burns out or a connection comes loose, people in the Control Room are on hand to replace or fix it (or call someone who can).  Since my job is to make sure that the software which reconstructs the particle tracks knows where all the detectors are, I’m more often in the CMS Centre.

The CMS Centre is meant to be an intermediate control room, where we can monitor data quality.  Tracks from charged particles passing through the detectors, showers from particles destroyed in the calorimeters, results of trigger decisions, symmetry and uniformity of the collisions events, all of these provide a first check that the data we are reading out make sense.  Specialists of each subdetector sit at stations around an oval room on the main CERN site, organized like the Enterprise bridge without a captain’s chair.  It’s pictured below on a day when we weren’t taking data (that’s why it’s empty).

The room was refurbished to house the Centre last year, but it’s in one of the oldest buildings at CERN.  Just outside looks like this:

The blue railing is one side of the “hallway” that leads to the CMS Centre.  In the middle of this picture are the glass doors that whoosh open when I wave my access card.  This building also contains the Proton Synchrotron (PS), CERN’s first accelerator (1959) and the first step in accelerating protons for the LHC.  It’s also used to store racks of old magnets and scanner tables.

(Scanners were people: in the days of bubble chambers, particle physicists hired them to scan through roles of slide photographs, projected down onto screens the size of kitchen tables, looking for rare decays.  It’s good that we use computers now— I’ve heard that the projections were bad on people’s eyes, and we expect to find 1 Higgs event in 1010 LHC collisions.)

Getting back to the CMS Centre, it’s designed to be something like a walk-in computer desktop.  The monitors are arranged in groups and multiscreen arrays, running without a window manager.  Each screen is a window, usually dedicated to only one control GUI or webpage.  I think that makes it easier for a group of people to look at the same data— flipping between windows is disorienting to everyone except the person at the controls.

The main display is on a projector for everyone in the room to see.  When I was there, it was cycling through several displays.  Here, it’s showing webcam pictures from the other CMS control rooms around the world: two at the CMS site (one is underground, the other on the surface) and the Remote Operations Center (ROC) at FermiLab, near Chicago.

In principle, people can monitor data anywhere in the world, but few experiments have ever been operated that way.  CMS has a large U.S. contribution, so FermiLab has dedicated a lot of floorspace to becoming a secondary nerve center for the experiment.  It will be interesting to see how well that works.

Before leaving, I’d like to show two more pictures that complete the ambiance.  Just down the hall, there’s a little coffee shop for getting espresso during a very limited fraction of the day.  Take note of the rose-magenta paneling and photomural of a flowery swamp on the walls.  When I first passed by this area in 2007, it was covered in inches of dust and strewn with stacks of sheet metal and broken desks— it obviously hadn’t been used since the decor was in style.  When the CMS Centre was built, someone evidently thought it was time to revive the coffee shop as well, restoring it to the time-warp splendor you see below.

Outside, the CERN site is an industrial park interspersed with grassy mounds with grazing sheep on top and accelerators underneath.  The roads are all named after famous physicists and the buildings are all named after numbers, not in any particular order.  Building 124 has a bicycle stuck through it.

I guess there’s nothing more I can say.

About these ads

Tags: ,

3 Responses to “Pictures from the Control Room”

  1. Garnet Hoyes Says:

    Very interesting. I am working at the Thai synchrotron, no where near as big as Cern but gives you the idea. See the web site and click on the machine status window.

  2. hina Says:

    000000

  3. http://tinyurl.com/hfccburne51890 Says:

    “Pictures from the Control Room The Everything Seminar” was a extremely excellent post, .
    Continue authoring and I’m going to keep browsing! Thanks a lot -Sang

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Follow

Get every new post delivered to your Inbox.

Join 102 other followers

%d bloggers like this: