Modelling Fukushima emissions and fallout

I’ve been reading with great interest a recently released approximation of the quantity and deposition of radionuclides from the Fukushima Dai-ichi nuclear power plant accident earlier this year. Recently published on the Atmospheric Chemistry and Physics open discussion forum, the PDF form (entitled “Xenon-133 and caesium-137 releases into the atmosphere from the Fukushima Dai-ichi nuclear power plant: determination of the source term, atmospheric dispersion, and deposition”) is freely available.

The discussion paper raises a number of interesting points, particularly the indications of pre-tsunami reactor containment failure as evidenced by radioxenon releases, the large contribution to overall emissions from the spent fuel pool fire (reactor building #4) and the upward rating on the overall estimated radionuclide emissions. I strongly recommend a read.

Photo credit: TEPCO

Exploding the legend

I am a bit late on this one given all the fireworks and general merriment of this month have past, none the less I feel the need to post a documentary the BBC did recreating the Guy Fawkes plot. (Background courtesy of Wikipedia for those not familiar with British history.) Except of course, they blow up their life-size replica Houses of Parliament. I didn’t record it when it was first shown 3/4 years ago so I was ecstatic when I found it on youtube, the quality is a little poor but it is phenomenal what that amount of gunpowder could have done.

Part 1:

Part 2:

Part 3:

Part 4:

Part 5:

Part 6:

Part 7:

Part 8:

Halloween special – Science of Candles

Welcome Wizards and Warlocks, something a bit special today (although I know I am a bit late with the Halloween reference) – my friend Louise pointed me towards this short youtube video explaining the science behind candles. It is rather interesting actually, so anyone with a love for science (or just pyromania) — enjoy 🙂

Distributed Computing : Folding @ Home Update

I used to be quite an avid F@H contributor, I ran several folding rigs and wrote a bit of code for addins to the project. I have not done any real folding since late 2007, that is until this week. I built myself an awesome gaming system last week containing an Intel Core 2 Duo 8400 CPU, 4Gb of RAM and a ATi 4870 GPU. I am extremely pleased with the way this system works and decided to try out some of the high performance clients the F@H project offers.

These essentially break down into SMP and GPU. The former, SMP (Symmetric MultiProcessing) is a client designed to spread the unit workload over the physical (or virtual) cores available on a system rather than using the single threaded normal clients. Disappointingly it doesn’t appear the cores have been modified to achieve this, instead it looks like a managed series of single threaded processes are used. I will need to investigate this further as I had some problems getting it to work – more on this hopefully at a later date. The latter takes advantage of the stream processors on the graphics cards to perform folding, it is in essence a great example of GPGPU. This client absolutely rocks on my ATi 4870’s 800 stream processors currently giving me an approximate PPD (points per day) of greater than 2200 using the Gromacs GPUv2 core.

So, to anyone interested in Folding at Home, despite not being on the ‘officially supported’ list, the ATi 4870 is a work unit crunching beast. 🙂

(Profile)