Scaw2006Day2

Even though I was feeling quite wasted I did manage to drag myself off to

Attach:spring_field_alternative_flyer.gif

which Jason tipped us off about and I'm glad I did because it was a really decent evening out. Danced to some music. Met up with people I hadn't seen in awhile (including ex-housemate, Sarah, would I hadn't seen in a really long while).

This afternoon was spent getting Qemu setup so I can test work things on Windows (yuck) although on the upside it may allow me to do some retro gaming as well. This evening we made dumplings and then watched Steamboy and Blues Brothers with Neil.

Perhaps the most annoying thing about the tutorial was that it was cut horribly short. According to the program it was meant to continue on until 17:10 but by 15:50 everyone was being asked to pack up and by that stage we'd only had about twenty minutes playing around on the ORION cluster. It was particularly frustrating for me since I'd gone to the trouble of dragging a spare desktop machine along in order to participate.

Rather that sit through four o'clock traffic on Friday afternoon Kevin, Neil and I adjourned to the Gardens Centre for coffee (and in my case breakfast).

The tutorial began with a walk through writing a simple MPI program to solve a 2D La'place equation in Fortran. Working through the problem went perhaps too slowly and it would have been nicer if the code had been a bit more organised (maybe some subroutines?) but it did manage to illustrate some of the basic MPI concepts (code executed depends on process rank, the need to exchange boundary information between processes, using the highlevel MPI calls like gathering and summing).

Perhaps the most annoying thing about the tutorial was that it was cut horribly short.

SCAW2006, Day 2

Friday was the second and final day of SCAW2006. Unfortunately I didn't make any of the morning talks (I have to work sometime :) but I did manage to attend the afternoons tutorial on a portable ORION cluster. Those who did attend the morning's talks mentioned Peter Kramer's (of SGI) one on future performance issues and Ian Reid's (of NAG) one on cluster benchmarking as particularly worthwhile.

SCAW2006, Day 2

9 April 2006

Friday was the second and final day of SCAW2006. Unfortunately I didn't make any of the morning talks (I have to work sometime :) but I did manage to attend the afternoons tutorial on a portable ORION cluster. Those who did attend the morning's talks mentioned Peter Kramer's (of SGI) one on future performance issues and Ian Reid's (of NAG) one on cluster benchmarking as particularly worthwhile.