Log in

No account? Create an account
Thesis Writing - Nick [entries|archive|friends|userinfo]

[ website | gagravarr.org ]
[ userinfo | livejournal userinfo ]
[ archive | journal archive ]

Thesis Writing [May. 30th, 2003|08:11 pm]
[mood |contemplativecontemplative]
[music |Sum 41 - Ride_The_Chariot_To_The_Devil]

As many of you will know, I'm a fourth year chemist. Being my masters year, this involves a year of pure research. None of this namby pamby lectures and exams thing, just pure research all year.

Unfortunately, this research lark comes with the need to write a thesis on my work. A 50-60 page (core text) thesis. A thesis that's due in in just under 2 weeks (12pm on Friday of 7th week to be precise). A consise, insightful report into my year's worth of (lack of) work. Fun.....

My work this year has been impressive in many ways. I've converted the group fortan code base to be explicitly typed, and fixed some bugs in doing it. I've written some MPI helper routines for the group code. These are cunning, in that they allow one version of the code to exist, and have it pick between single threaded and MPI at compile time. They also keep all the pesky, hard, multi threaded stuff away from most of the hard science, so should allow most people in the group to use it. The type of parellisation used is a rare one, of job parellelisation. This takes advantage of the fact that we need to run a large number of jobs at similar state points, and also repeat runs for averaging. As such, it uses mpi to handle co-ordinating the handing out of job parameters between individual threads on a number of machines. While some perl scripts on an OpenMosix system could do similar, the OSC provides MPI routines and not OpenMosix facilities, so we've used them.

Almost none of this can make it into my thesis. A small bit is in the appendix.

Armed with this shiny MPI code and an OSC account, I've produced mind boggling amounts of data. About 120 gigs worth to be precise, which is why I had to go and buy a new disk for my machine.

Writing up 120gb worth of data is interesting. Most part IIs (4th year chemists) don't have enough data (eg Hilary). I have so much that I've barely started analysing it. One of next year's part IIs could write a thesis analysing the rest of my data.....

Currently, the thesis stands at 33 real sides (49 sides with TOCs, appendicies, bibliographies, summaries etc). Phew....

I'm aiming to have a rough draft done by wednesday. I was aiming to have it done by today, but that hasn't happened. There are still a few sections to flesh out, conclusions to write, a summary to write, and a whole section (on cation-cation double peaks) to write or ditch.

Still, people who've read it think it's quite good, which is a bonus.

A side effect of all this is that I haven't spent nearly as much time chilling as I might've liked. Ah well, the price of not starting writing earlier, and slacking for much of the year....

Hopefully it'll all be done and handed in early. If it isn't, I'll be looking kinda stressed come the end of the week after next!