The week started off quite sadly, with a funeral for my cousin's husband on Monday. This was quite difficult, since we have been friends with both my cousin and her husband (who was about 12 years older than me) and one of my cousin's kids, who is about 12 years younger than me. He died of brain cancer. It was a sad day.
Tuesday, my mom decided to steal the crablets for a few days. So in the AM we packed and ran some errands, then had lunch, played with the neighbors (I worked), and said goodbye. I wend to work for a few hours, then the Mrs. and I caught a movie (Hancock - I would give it a 4 out of 5). Home, eat some blueberry pie, and bed.
Wednesday, Mrs. Crab and I biked to work. Since it was two weeks since I had ridden, my legs burned and burned and burned - and that was before I got out of the driveway. For me, Wednesdays are meeting day. At 2:00, I finished the morning meetings, and headed out for lunch.
We were given the opportunity to run on a brand new super computer that is owned by the National Center for Atmospheric Research. It has 4096 processors, and we are going to be allowed to utilized a significant portion of the machine for long periods. In order to prove that we can really utilize the machine efficiently, we had to do a bunch of tests. So, Wednesday was spent trying to get the code working on the new machine and doing "scaling curves" - this is where we create a problem, run it on 32 processors, then 64 (which should be completed twice as fast), then 128 (4 times as fast), 256 (8 times as fast), etc. If the code is actually 2, 4, 8, etc. times as fast, it is said to scale perfectly. It is impossible to do forever, since you run out of cells or whatever. For us, given the problem size that I chose, we ran just fine up until 1024 nodes. I then redid the calculation for a problem 8 times larger, and showed that we could run just fine up until 2048 processors (1/2 the super computer!)
I biked to Mama-crab's office, home, cereal for supper (we eat so well when we work long days), a little "So you think you can dance" (I am ashamed), and then bed. Crab-kids doing fine at grandma's. Playing at the beach.
This morning I took the dog (scaredy-crab) to work, since I was at work for a long time today. More scaling and testing at work. I had an annual report due a tiny bit ago, so I had to write about what I was working on for the last year. I love reports. They are so fun. The super amazing thing about the National Science Foundation is that I finished my annual report today, submitted it this afternoon, even got a question on the science from my program manager, then, just now, I got the money for next year. (time line - reminder at 7:33 AM, turned in at 1:47 PM, approval at 2:00 PM, new money at 9:16 PM.) I am not going to dog on any other funding agency, but NSF has it's sh*t together.
Tomorrow we are off to Grammy-Crab's to watch her in a play and to pick up the kids.
Well, I would say that this is a typically week, but, I honestly have no idea what a typical week looks like. So, I guess it is typically in that it is completely unlike any other week I have had in the last many, many months.
1 week ago