Hi my name's Alistair Blachford and I'm the bioscience computing manager. Since 1985 I've been working here. I've been paid to help a whole bunch of really smart people figure out how the living world works. I came to UBC in the fall of 1980 to do a masters with Carl Walters and I thought UBC was the coolest place ever because we had an apple orchard. We had about six or eight of these (The Apple II) spread around the periphery of the Ralph York room. It's on these so we did all the simulation assignments and everything for Carl's courses and later on when I joined the staff of the computing unit, it was called something different back then, all our computing was done on a great big machine called the vax 11750. Our interface for that was just dumb terminals that would send characters back and forth from the machine. Everybody used the one machine and then the architecture changed so that people could have more computer power on their desk.

The PC revolution the IBM PCs came along and the Apple macintosh came along which put a ton of power on a desk, so all of a sudden people could be master of their very own computer power and the way we had to architect the system that the researchers used changed. We changed from central machine to client server, so I as manager got us through that transition. Don Ludwig contributed a sun workstation which again was meant as a desktop thing, but it's a very powerful one and I did a little presentation at a faculty meeting and I said we got to switch architectures because this vax thing is going to go extinct and it's too expensive. A bunch of faculty members chipped in PCs to be clients and we basically set up the first grad lab type thing. Judy Myers and Jamie Smith threw in machines that we called bug and bird and we had the other machine as a server. That was our transition to client server, the way things are now mostly and computing power just keeps escalating at a mind-boggling rate as I think I can show with the little computer museum. As the computing power increased the kind of kinds of questions that could be asked changed also. The cute little questions remained, but you could ask really big questions, do meta-analyses use big data and do all the genomic stuff.

The genomic stuff is very computer intensive and RAM intensive. I would never have imagined that we'd have multiple machines with one and a half terabytes of RAM and probably in a year or two that's going to seem small, but that's what you need to put together your genomics data. Most sequencing is still shattering the genome and then getting the computer to line everything up. That was unthinkable a bit over a decade ago, I'd say 15 years ago. Now although we have all these massive computers on our desk compared to old times, we need even more massive central computers. We're using both levels now the desktop power and the central power in our departmental machine room, but we're moving much more in into an era where we're going to be running everything in virtual machines. We're going to need to do that for scientific research to to kind of wrap up and containerize our work on each project.

Right now the problem is you can store your data, but if you ever want to use it 10 years down the road all the software that you use for it has changed. I think in the future the the zoology computing unit is going to offer um a way to do your work on a particular project in a virtual machine that you can just tie it with a bow: data, software, everything and park it there. That way you can achieve a far better way to replicate results than we have with the way we do analyses right now. That's a future trend and week by week the power is going to go up and we're going to shake our head week after week not believing what power comes along.