Man versus Machine, Part Two
Wed, 11 Apr 2012 04:50:00 GMT
The second part of a two-part interview with Professor Dave Cliff
In this second part of a two-part interview for the High Frequency Trading Review, Mike O’Hara concludes his conversation with Professor Dave Cliff of the University of Bristol. Professor Cliff is Director of the UK Large-Scale Complex IT Systems Research Initiative, and a member of the Lead Expert Group of the UK Government’s Foresight Project on The Future of Computer Trading in Financial Markets.
The first part of the interview can be found here.
HFT Review: Dave, welcome back. In the first part of this interview, we discussed the background of computer-based trading and the use of artificial intelligence and machine learning in financial trading systems and we looked at how humans and machines might interact in the financial markets of the future. Now I’d like to get your insight on some of the emerging technologies that are starting to become prevalent, starting with cloud computing.
Dave Cliff: Cloud has gone from being a phrase that nobody knew, to a kind of bandwagon buzz word, to something that is now obviously here to stay.
If you look at the history of commercial provisioning of ICT, roughly once every ten or twelve years there’s a major shift. Way back in the dawn of the industry, if you wanted a computer, you had to buy what was essentially a single user mainframe from IBM. The dominance of the mainframe lasted roughly a decade and then IBM’s incumbent position was threatened by new arrivals, new companies that saw new opportunities. In particular there was this small start-up company called Digital Equipment Corporation (DEC), who realized you could take everything a mainframe was doing, but using new technology in both hardware and software, you could deliver that for much less cost and in a much more manageable form. DEC did very well with the dominance of mini computing for about a decade, but then along came micro computing, i.e. single board computers, based on single chip microprocessors, one per desk. That was very disruptive and threatened both IBM and Digital, when companies like Compaq (who eventually bought Digital) rose to prominence. Then people realized that you could economically network the PCs together to get local area networks, client-server architectures, that kind of thing, and that lasted for about a decade.
HFTR: Then along came the Internet…
DC: Actually, the really major evolution was the Web, not so much the Internet. The Internet had been around for 15 or more years before the Web took off - the Web was the recognition that if you have all these local area networks and they can talk to each other, then in principle anyone can talk to anyone, any computer can talk to any computer anywhere in the world, so long as you agree the protocols: http and html.
So we’re due one of those big revolutions in ICT and I’m happy to bet a couple of bottles of champagne that cloud is the next one because it offers the opportunity to dramatically reduce the cost of computing. Interestingly it means that what was previously a capital expenditure item becomes an operational expense. You only pay for that which you need, so you don’t have to worry about dormancy.
HFTR: How relevant is cloud to high frequency trading?
If you are not a registered user, click here to register for free