|
|
|
October 23rd, 2009
05:37 pm - Return to Oak Ridge
I just got back from the lab a few minutes ago. I'm pretty tired. A 2.5 hour long meeting of the research group in which I had nothing to say because it was all about two codes I'm not involved with really takes it out of you.
|
October 19th, 2009
05:06 pm - Scientific Linux
We're getting a new node cluster here in the department, and I need to be able to make sure it be able to compile the software we do our calculations with. It'll use Scientific Linux. So I went to run Scientific Linux in VMWare so I could test it out/try compiling things in it. The Scientific Linux installer refused to notice the virtual harddrive that VMWare created for it.
Sciencefail. :(
|
September 14th, 2009
04:54 pm - Bader-Deufelhard
I just love the names of some of the algorithms I get to work with. This one, I think, wins because it sounds like the name of a Nazi general defending the Rhine with his division of Hitler Jugend against overwhelming Allied opposition.
It's actually kind of a neat method, and according to a paper by Timmes that I read it compares favorably against most other approaches that people have tried to integrate nuclear reaction networks with.
|
September 9th, 2009
04:08 pm - Committee Update
Just got done giving my committee their annual required update. I think it went well enough.
|
August 24th, 2009
11:55 pm - Quick update for my science groupies.
It looks like the last problem running on the supercomputer is out of the way. Hopefully I can get back to my thesis...
|
August 13th, 2009
07:56 pm - Maybe...
I think maybe I've slain the supercomputing beast, finally. I'm not 100% sure yet, but I'm hopeful...
|
August 4th, 2009
11:46 pm - Frustration
It's frustrating when no matter how many angry faces you make, and in all defiance of how C is supposed to work, the supercomputer you're trying to run your code on refuses to assign the address of a variable to a pointer...
|
August 3rd, 2009
11:03 pm - Research
So today was a bad day for science. But I think right before I left work, I figured out what I need to do to make tomorrow a better day!
|
July 28th, 2009
07:00 pm - Science progresses
As always. More bugs squashed. More spotted. It's a vicious cycle.
|
July 9th, 2009
03:19 am - Whew.
Sat through two masters thesis talks yesterday. That kind of stuff wears you out.
|
June 20th, 2009
04:11 am - Let me regale you with tales from the Lab
Today I went to ORNL for the first time in a long time. I get there, my badge has expired. I go to get a new one: they're shutting down the servers that hold employee information in 20 minutes. I get my badge and go to register on the network, of course I can't because it can't look up my badge number. Then there's a meeting, time spent doing nothing, hair pulling over how Eugene is configured, and then I finally get to go home.
It was less exciting than all that, really.
|
April 8th, 2009
07:55 am - MATLAB
So the installation process for MATLAB was a pain. It took me like an hour and a half. Mainly because the directions they gave me didn't coincide with what actually happened. The directions assumed somehow that the install script would be invoked automatically by the download script, but it wasn't. I had to figure out which file the install script was inside of, then untar it (not that was named .tar to make that easy to figure out), then finally I could invoke the install script. Kind of a pain.
Also these companies and their draconian licensing schemes are ridiculous.
I think it would have been easier for me to get an illegal copy of this stuff and install it... At least pirates give you directions that work.
|
April 7th, 2009
10:17 am - Mmm Maple
I just installed Maple on Andaman in my office. Soon, all of its plotting worksheet-based power will be mine to control!
MUAHAHAHAH
|
March 30th, 2009
01:25 am - Silicon Burning
Definition 1: The process by which stars burn Silicon to generate Iron-group elements.
Definition 2: What happens when you try to simulate the process of turning Silicon into Iron and Nickel. See below the cut for evidence supporting this definition.
( Read more... )
|
March 29th, 2009
11:29 pm - sigh
Don't you hate it when you forget something that's kinda important?
|
March 16th, 2009
10:17 am - geat: no longer exploded
Brad came up from the electronic dungeon of doom and in about five minutes geat was working. It was the ram in it that had gone bad. And apparently my earlier ramswap didn't work because the RAM requires you to try to break the motherboard before it will seat properly.
The more you know.
|
March 13th, 2009
08:21 am - Science: up to 11
So yesterday, we cranked science up so high that we blew a circuit breaker.
How?
With my computer.
So over the weekend geat went nutso, and on Tuesday while I was going to fix its configuration and get some things that didn't load running again, the master node decided it was time to go belly up and wouldn't reboot. At all. So I brought my old(er) computer up to the office and got it all set up to be able to run our calculations and things. Well, yesterday is when we hit critical mass. The combination of knowledge, bug fixes, debugging, and the sheer awesome that having me in a room generates was entirely too much for the machines. Early in the afternoon we blew one of the breakers in my office. Specifically it was the breaker that my computers were hooked up to.
The synergy of me, with two of my computers, in one space was simply too much for the universe and in an attempt to prevent space-time from curling back upon itself and collapsing into a singularity the hand of fate reached down and flipped a breaker, killing my two calculations and plunging the room into darkness.
We didn't let this stop us, of course. The awesome maintenance dude came up and we were rocking again about thirty minutes later. And what kind of rocking were we doing, you might ask?
I'm glad you did.
It was this rocking.
Thermonuclear burning. Fast. Easy. And without any bizarre crashes or bugs getting in the way of the calculation.
Later in the day Dr. Guidry basically reran Viktor's thesis, the director's cut, and it's awesome too. But I'm not putting that graph here. It might be worth a paper or something.
Blown breakers, blown up stars, and me.
Our science goes up to 11.
|
March 12th, 2009
11:54 pm - Science as of this afternoon:
 Quickpost this image to Myspace, Digg, Facebook, and others!
|
March 10th, 2009
09:30 pm - Geat 3: The Reckoning
Today, I went to restart geat after bringing the down'd nodes up, in the hopes it would run mpdboot and everything would be kosher. geat's master node didn't come back up. Instead it didn't get past the POST. I had to go interpret beep codes and LED lights. Something wrong with the RAM. So we swapped some RAM between nodes and it still didn't work. So the masternode's memory controller must be borked. So there went 2 hours. Then I spent like 4 setting up my (older) computer in the office. What a waste of time. I'll take geat's master node down to the electronic shop and hope they can fix it, or cannibalize another node to do the job.
|
January 7th, 2009
02:54 pm - Argh. Frustrating
Back in December I did a bunch of calculations.
Now I can't find the data from them.
>.
|
|
|
