In investigating the crash, Mr. Berman says he finds himself in a position similar to his physics work 20 years ago, when he was collecting huge amounts of data and comparing the competing views of many laboratories on a question dividing particle physics — whether the neutrino, one of the least known and most common elementary particles, actually had mass.

Particle physicists sometimes show up in surprising places.

This essentially shows the amount of proton collisions created by the Large Hadron Collider as a function of time.

I have my own personal “analysis” code set up so that every Friday I run on the newest data available.  Then, by the afternoon or Monday morning I have new plots to look at and show colleagues.

I do this on Friday because that’s when a list of “good” data-taking periods is published.  That is, there is a group of people that decide when the detector was fully operational, working as expected, and recording useful data.  They then publish a list of this “good data”, and everyone else uses that list so that they run on useful data.  An example of “not useful” data (for physics) would be if one or more parts of the detector were off, or not working properly.

The plot I show here is the amount of proton-collision data available.  It grows more than linearly because the intensity of the  beams is being increased as well.  Someday, in the distant future, the LHC will be able to deliver an amount of collisions in one work-week equivalent to everything we’ve recorded since turning on in May this year.  Someday.

Even in my hometown of Madison, WI there is a French bakery owned by a french family and they close up shop for most of August.

The disappearing of French workers also happens at CERN – professors, scientists, etc, many of them are gone.  That leaves the rest of us with the chance to either get ahead in our work, or relax and take it easy as well.

(Oh, and did I mention that the French also have a 35-hour work week?^[3])

Don’t worry though, the LHC is still on and they’re trying to reach higher beam luminosities.  At the moment they’re working on some cryo problems:

The energy of the proton beams is in black and goes with the right-hand y-axis, “Energy (GeV)”.  As I write this the protons are around 500-some GeV and being ramped up to 3500 GeV which should take about half an hour.

Once both beams are at 3500 GeV and they declare stable beams, it’s time to record some data with the most bunches in the LHC to date!

Right now the luminosity is at 10³⁰/cm²s.  Up until now, the luminosity collected these past few months has been at a luminosity around, at most, 10²⁹/cm²s.

This is significant because this basically means that in about a day we can collect the same amount of data that we have collected over the past few months.  That’s why it’s so useful to do studies on increasing the luminosity rather than continuing to run at lower luminosities.

There’s a short term downside though: doing studies to increase luminosity makes it hard to get clean, stable beams for data taking.  It’s kind of like deciding whether you should buy a computer now, or wait a few months until prices come down and hardware is better.

The LHC has a balanced program of stable running and also doing studies to increase luminosity.  From the looks of it, it’s going well.

Vending machines do not take more than once kind of currency.  Also, border guards don’t take kindly to bringing wine or meat across the border.

As for the American dollars, I only happened to have those because I recently traded euros to someone who was moving to France permanently, while I was going back to the US within a week or so.

Proton-Proton collision simulation "jobs" for the CMS detector running on the grid.

To compare with the data we record from our detector (CMS), we need to run a few simulations…well more like billions of simulations.

Each “job” in the plot above is actually a program running on a computer at a university.  Each program typically simulates a few hundred, or a few thousand, proton-proton collisions.  Each individual “collision simulation” calculates what a certain kind of collision would look like in our 12,500-ton detector.

And I don’t mean they just make pretty pictures.  A single simulation really consists of: some particles within each proton interact with some probability, they produce other particles with some probability, those particles decay to other particles with some probability, and so on…  Eventually, stable particles are made and the passage of those particles through the detector are also simulated.

As you can imagine, this requires a lot of random numbers.  One mistake that happens sometimes is that different jobs have the same initial ‘seed’ for the random numbers, and this results in duplication of simulations.  Not only is that a waste of CPU-cycles, but it also means a fuller range of collision possibilities doesn’t get simulated.

My job at times is to herd thousands of simulation jobs at a time to various places and monitor them, make sure they don’t crash, and finish in a timely fashion to return the needed data.

By the way, when I wrote the job monitoring script that makes plots like the one above (written in Python and using matplotlib), I tried using their school colors when I could, but sometimes that resulted in colors that were too similar or confusing.

This fall our 3-year grant is up for review, and he’ll help decide our fate, basically.

Our group had about 9 graduate students there and he asked questions to figure out what problems we were experiencing either within our group, within particle physics, or living in Europe.

Towards the end he also asked us about what we all wanted to do after we graduate.  He then led us through a somewhat sad “back of the envelope” calculation:

“Lets say the average professor’s tenure at a university is 30
years, roughly.  That typical professor has about 2 graduate students
at any time, and the average time for completion is 6 years.
So, the typical professor produces a total of about 10 PhD’s.
Well, they only need 1 to replicate themselves, and 1 more to
replicate positions available at national labs.  And that’s it, that’s
all there is room for in academia, typically, 2 out of 10.”

It’s an over-simplified example, but I think not too far off the mark.  About 1,000-some physics PhD’s are awarded in the US every year(AIP), but the number of vacant positions at universities each year is only a fraction of that(AIP Chart).

Update June 13:  I began searching for the names of my advisor’s former students and happened upon an on-topic article from the American Physical Society, Sean Mattingly, PhD High-Energy Particle Physics, Dedicated Client Support, Bank of America.  Sean is quoted as saying “I think every student should be thinking about a job outside physics.”  And that “in grad school we all think that we’re on the academic path, but you’re not – there’s a lot of competition for the few jobs available and most of you are going to have to leave the field.”

We were so close to getting to record some more collision data during this shift, only for the beam to be dumped due to some problems.  I’ll have to ask experts to see what a “1/3 resonance” means.

Hopefully they’ll have more luck during the morning and afternoon shift so that by the time I’m on shift again at 11pm I’ll get to watch more data being taken in person.

So, I’ll keep watching the trigger recording cosmic ray muons, making sure things stay looking ok.

In the meantime I can keep working on a paper we’d like to have a draft of by Monday.

For example, this means that if a professor doesn’t want to do shifts, their scientists, post-docs, or students must do them.

One complicating factor is that not every shift is worth the same.  The least popular shifts, or the ones “harder” to do – like overnight shifts – are worth more than others.

Here’s how many points each shift is worth in our collaboration:

• Weekday morning shift (7am-3pm) is 0.75 points
• Weekday afternoon shift (3pm-11pm) is 0.75 points, and
• Weekday overnight shift (11pm-7am) is 1.5 points.
• Weekend shifts add extra +0.5 points to above.

And since we are asked to do 24-points worth of shifts in a year, what kind of shift is most attractive to me, an unmarried, childless, young graduate student?

The weekend-overnight shifts, of course!  At 2-points each they’re pretty attractive.

Sometimes you have to take the shifts you can get, however, so I’ll be do weekday overnight shifts starting tomorrow.

Control room for the CMS detector.

]]> http://www.quantumdiaries.org/2010/05/30/scoring-points/feed/ 2 http://www.quantumdiaries.org/2010/05/17/back-to-europe-and-flying-comfortably/ http://www.quantumdiaries.org/2010/05/17/back-to-europe-and-flying-comfortably/#comments Mon, 17 May 2010 06:01:49 +0000 Mike Anderson http://blogs.uslhc.us/?p=4718 After a short stay in the US, I have returned back to CERN.

I flew on a Boeing 767 for about 9 hours.  That was not a comfortable flight.  An Airbus A330 has much more leg room.  Also, on the Airbus, each seat has it’s own personal screen to select from a couple dozen movies and shows to watch, whenever you want.

The science of a nicer airline flight isn't so complicated.

Airplanes could be more comfortable, but there are things you can do yourself to make trips better.  The best thing I ever did to make all my flights more comfortable was to buy noise-canceling headphones.

I was skeptical of noise-canceling headphones at first.  I didn’t have any friends that owned a pair, and most stores don’t have them out in the open for you to test.  I did research on them, all the way from the basic physics of noise-canceling headphones, to actual reviews of headphones.

Basically, noise-canceling headphones reduce low frequencies the most.  That means engine noise, low rumbles, wind, etc are reduced considerably, but you’ll still be able to hear a baby cry, cat meow, or a phone ring.  (I hope you never have to sit next to a cat on a flight.  I love cats, but I’ve found they’re not so pleasant on airplanes.)

I’ve been using noise-canceling headphones for two years now on all my flights, and I can’t recommend them enough for people that fly a lot.

How would have Richard Feynman or Enrico Fermi felt about having to use PowerPoint every week?

High-energy particle physicists involved with CERN experiments spend a lot of their time creating, editing, and reading three types of things: emails, computer code (C++ & Python), and PowerPoint slides.

The New York Times recently had an article about PowerPoint use by the military in an article called, “We Have Met the Enemy and He Is PowerPoint.”  It quotes several people in the military who are quite sick of the program.  One general is quoted as saying “PowerPoint makes us stupid,” and another says, “It’s dangerous because…some problems in the world are not bullet-izable.”

The article also references a popular essay by a retired Marine colonel, “Dumb-dumb bullets” which describes in detail why the author believes that “PowerPoint is not a neutral tool — it is actively hostile to thoughtful decision-making.”

Ouch.

There were hundreds of comments on PowerPoint after that NyTimes article that I’ve only read a sample of.  It’s interesting to hear other people’s experience, because PowerPoint is such a big part of life in particle physics.

I’ll avoid mentioning how I personally feel about PowerPoint for now, but let me describe its use within high energy physics (along with Apple’s Keynote).

As a graduate student working for a detector at CERN (specifically: CMS), I attend about 4 meetings a week.  These meetings are attended by a dozen or so people and are usually 1 to 2 hours long with, on average, about 5 PowerPoint presentations.

Click to englarge. Length of PowerPoint presentations from a sample of 25 physics meetings in Spring 2010 for the CMS detector. "Backup" slides were not counted. Bins inclue their lower-bound (i.e. a 5-slide talk goes in the "5-10" bin).

Thus, every week I spend about 6 hours a week in meetings watching PowerPoint presentations – and I know this puts me on the low end of a particle physicists’ “meeting-attendance” distribution.

Our collaboration has about 1,500 physicists – is it the best use of their time for each of them to be spending spending several hours a week watching PowerPoint presentations?  Management doesn’t seem to think so – they’ve created a task force to try and reduce the number of  meetings.

But there’s a related problem: the time spent in the existing meetings is already too high.  This is partly the result of the long tail in the graph shown here.  It shows the size distribution of a representative sample of presentations.  Most people hit a sweet-spot of 5 to 15 slides.  What about the talks longer than that?  Are those presenters making effective use of people’s time?  Or could they summarize their work more effectively?

I’ll put forward the claim that the long tail in that graph is not from most people occasionally making a long talk – but rather, it is from a few people who consistently make long talks.  They should be encouraged to cut slides, summarize, and focus on showing only the most important plots.  (Other ways to reduce the burden of meeting were given by another blogger here in the post “I hate meetings.”)

So yes, physicists are dependent on PowerPoint, it is simply the way they share information in a meeting.  So much time is spent watching PowerPoint presentations that it’s easy to say, in general, reducing the length of most PowerPoint presentations would benefit everyone.  Put details in the back-up slides – it would help put an incredible amount of human hours towards more productive activities.

Click for live LHC status.

The LHC is continuing studies to reach higher beam intensities.  There was a “technical stop” the past few days to turn on more of the machine protection system so that the they can safely operate at higher intensities.

A 3.5 TeV beam of protons can cause plenty of damage if there’s uncontrolled beam loss!

My office at CERN.

I’m curious to know what the best-looking office is at CERN. (If you know someone or if you think your office would be in the running for that let me know and I can post a picture!)

From what I’ve personally seen, most offices here aren’t all that different from mine.  A little cramped, but at least they have big windows.  The plant was my own doing – and it is still a little pathetic now due to a lack of watering over the holidays in December.

Looking at that picture now it just occurred to me you would notice we use mac laptops (mine is plugged into a monitor).  The use of macs among high energy physicists is pretty high – higher than the overall population – and I wonder how it became that way?

–MikeA