Matt Lebofsky

Information Systems Analyst IV
SETI/Space Astrophysics Group
Space Sciences Laboratory
7 Gauss Way University of California at Berkeley

email: mattl at ssl dot berkeley dot edu

About Me:

I'm a general computer geek here at the Space Sciences Laboratory, on and off for about 20 years now, spending most of that time working on several SETI projects, including SETI@home, SERENDIP, and BOINC. More details about my work are in the Frequenty Asked Questions below.

I am also a professional musician. Weird, adventurous stuff, mostly. I've played hundreds of shows in dozens of countries. More information about all that can be found on my personal web site.

For those entertained by such things, here's a (likely outdated and incomplete) resume.

I sometimes also rant about the SETI@home servers issues du jour in the SETI@home technical news forum.

Frequenty Asked Questions:

Q: So... have you found anything yet?

A: Nope.

Q: Well, would you even tell us if you did?

A: Of course! Why wouldn't we? This project is largely funded by public donations, not some big, evil government conspiracy. Plus we'd rather collect our Nobel prizes instead of keeping things secret.

Q: So what exactly is it that you do, Matt?

A: The UC Berkeley SETI program with its several disparate sub-projects share the same computing facilities and tiny staff (about 5 FTEs at any given time, if that). Our skills greatly overlap, but here are my general task categories:

Systems Administration: I help manage the whole SETI network of about 25 very busy machines, pretty much all linux at this point. Added together my server racks contain roughly 200 CPUs, 700GB of RAM, and 400TB of disk storage that I have to keep up and running 24/7 and secure.

Webmaster: I have several web servers, running either apache or nginx, each getting millions of hits per day. I deal with a lot of load balancing, CGI magic, php scripting, the usual LAMP stuff.

Database Administrator: We have several databases running a mixture of mysql and informix. The mysql databases are quite active, millions of rows peaking at 30000 queries/sec. The informix databases are quite large: hundreds of billions of rows. We spend a lot of time adjusting these databases to fit our shifting scientific efforts.

Network Administrator: Our SETI@home project put us on the map as the world's largest volunteer computing project, with hundreds of thousands of computers around the world processing our data and accessing our servers 24/7. And thus we have to deal with some world size problems.

Scientific Programmer: When I'm not occupied with the above, I get to write some code, crunch some numbers, and look at pretty graphs. I'm usually coding in lot of C, pretty much all the common scripting languages, gnuplot, idl...

The long and the short of it is that while I get to look for little green men, I'm largely doing the same mundane stuff most people in IT do, as well as some basic administration tasks since we have no such staff (writing fundraising letters, generating grant proposals, general public outreach, etc.).

For more nerdy details, you can sift through many old technical news updates I made on the SETI@home site over the years.

Q: How did you get a job working for SETI?

A: The presumption, of course, is that working for SETI is some kind of elite, exciting, glamorous job for which only the most brilliant nerds qualify and requires some kind of rigorous security clearance, or mental stability test, or whatever. This isn't the case at all - it's just a garden-variety data analysis project just like any of the zillions at universities all over the world. Anyway.. maybe this will clear things up.

I was a computer geek of the nth degree as a young man, steeped in machine language coding every night for hours on end, but kinda got bored with it all during college. I also got disillusioned by how the world was adopting Windows more and more, while my beloved Amiga was falling by the wayside. So I wasn't exactly seeking work in the field with much verve by the end of my senior year.

Shortly after graduating I moved to California with absolutely no friends, no money, and no plans whatsoever. I'm a huge proponent of "making your luck," i.e. throwing myself head first into random, difficult situations and seeing what happens. Finding steady work was hard. I was competing with MIT and Stanford nerds for pre-dot-com jobs so that went nowhere. After random stints in various offices and six hellish months in the ad business I found myself jobless yet again. I applied as a temporary employee at Berkeley just to get some quick bucks. I never heard back, so I followed up with a call two weeks later. I misread the phone book entry and mistakenly called the director of temporary services himself, and he actually answered his phone. Luckily he wasn't busy and proceeded to look up my application. Seeing I had computer skills, he said a position just opened at the Space Science Lab that morning. I lied and said I was an expert at troff/LaTeX. I went to work the following Monday. Moral of the story: it's easy to get your foot in the door as a temporary employee, especially if you accidentally call the wrong number and then exaggerate about your skills.

I became a member of the Space Astrophysics Group (a.k.a SAG) which was a set of disparate projects that shared scientists and staff. The main projects at the time were EUVE, EUVIP, ORFEUS, and SERENDIP (a.k.a SETI). I held a mostly administrative role - I helped format papers and proposals, monitor/database purchasing and accounts, and help one completely computer illiterate scientist send/receive e-mails. Not very exciting, but it was steady work. This was February 1994.

A mere few months later my supervisor up and quit one day. It's a long story - let's just say the circumstances were unusual, and these circumstances include the words "Venezuela," "surgery," and "ten thousand dollars." I was promoted into her spot and became a full-time employee. I also got her office. Sweet.

One day was particularly dull so I started snooping around the network - nothing malicious, just curiosity. Jeff Cobb (SERENDIP programmer and SAG systems administrator at the time) immediately noticed me issuing "sudo" commands on his servers. Oops. He asked if I knew what I was doing, and once aware of my former life as a computer geek he asked for my assistance doing network backups and software installs. I accepted these tasks, and even got called on to do some SETI-specific analysis chores. Moral of the story: snooping around your own network is a great way to flaunt your abilities and interests to your supervisors and get more fun stuff to do.

Due to a huge oversight I managed to have one of the best offices in the lab for over a year. That is, until the director of the ORFEUS project (Dr. Mark Hurwitz) came in one day with measuring tape saying, "don't mind me." He then proceeded to take notes about the room's dimensions. Within a couple weeks I found myself displaced into the carrels next to the printer and fax machine. Mark is a really nice and brilliant guy, so I didn't begrudge him his own office, which he certainly deserved more than I. However, I was perfectly healthy for years up to this point and now my new desk was situated under this giant vent. Over the next six months I got an equal number of nasty sinus infections. I demanded a building inspector come and check out this vent. Despite my recent health record and the layers of dust and dead flies on my desk, the inspector insisted my chronic sinus infections must be due to something else. I was sufficiently annoyed by this and the lack of career advancement opportunities so I quit. That was December 1996.

I enjoyed a half year off working on music, touring the country, and slowly draining my bank account. During that time I offered infrequent yet free tech support to my former lab workmates. Right around the time I ran out of money they called asking me to come back. I said I would if I could be rehired as a Programmer/Analyst and work flexible hours. They agreed. Moral of the story: quitting a job is a great way to get a long vacation, followed by a handsome raise and promotion.

Now that I was doing primarily systems stuff, I was working directly with Jeff, sharing lab space with him and SERENDIP director Dan Werthimer. Due to proximity I ended up primarily working on SETI, helping to wrap up the final data analysis of the SERENDIP III project and ramp up SERENDIP IV. Meanwhile SETI@home was coming into being (Summer 1997). The remaining SAG projects (and their funds) were waning, so they didn't mind so much that all of my time was being spent on SETI. By the end of 1998, the SETI@home buzz generated enough funding to hire me full time, and that's basically the whole story.