Power Play

james-rundle
Shaving milliseconds off milliseconds.

Earlier this week, I was having a coffee with the CEO of a large vendor at a bar in Bank, the center of London's financial district.

"We're finally starting to break away from Microsoft," they said to me, and proceeded to outline all of the reasons why Bill Gates had stifled technical development over the past 20 years. An almost-evangelical zeal came over when they began to talk about cloud and mobility.

"You know, the future is really in that," they said. "But taking it to the next step, as in, you'll be able to look in the mirror and call up a display to do something quickly, then exit it and it's just a mirror again. Or have a device that ionizes the air in front of you so that you can have a projected display. It's all about points of consumption."

Steam Shaft
They explained how various other technologies had developed through the years, taking the idea of power as a key example. With steam power, for instance, a central stack would power the factory, with the shops on either side of it, moving up through the building. As electricity developed, the model remained until people realized that they could plug in, and so, as a result, construction developed on a lateral basis. You didn't need these enormous verticals that centered around the energy any more, the energy moved to fit with the design and purpose of the machines it powered.

Microsoft, obviously, was the steam shaft in this allegory. The design of Windows systems had inhibited development along one path, and tying up with Intel, who they described as doing a very good job broadly, had created a situation whereby systems and processes such as low-latency had to work around the framework, rather than being supported by it. With the advent of Linux and more commonly-accessible distributed computing, outside of the mainframe world, this has begun to change.

With technological phenomena such as Big Data exerting pressure, solutions outside of what may have been considered traditionally "financial" in the strictest sense are being pursued with vigor.

Power
One of the key points of the meeting was around IBM's Power system, and its applicability for financial services. Most systems run x86 through Intel, with a structure based around a central hub where information is passed in and out. The hops cause latency accumulation, which can't be avoided due to the nature of the build. Power runs without these, he explained.

They had been extolling the virtues of the system to various financial institutions for some time, but for various reasons, the project had never taken off. For market data, low-latency operations, matching engines, they seemed to have examples of how they could have dramatically improved performance outside of the norm. The person in question seemed to believe that they'd finally reached a point where big test cases would emerge, however, although I can't mention the specifics of what was discussed just yet.

With technological phenomena such as Big Data exerting pressure, solutions outside of what may have been considered traditionally "financial" in the strictest sense are being pursued with vigor. Take Hadoop solutions, for instance, or DataArt's support of open source initiatives.

Given how much used to be produced on a proprietary basis, in-house, did anyone realistically think that backbone systems would be customized from publicly-available source code today?

Stranglehold
I'm not sure that I entirely agree that Microsoft and Intel have strangled tech development, if I'm being completely honest. I suspect the conversation on one side was coming from someone with a rich background in software engineering, talking to someone with a degree in politics and a career in journalism, and so the point may have been lost in translation. But one question I asked resulted in an answer that both of us fully appreciated and ties in to why extraneous solutions are being sought out.

"If Power is everything it's cracked up to be," I said, "then why isn't everyone using it?"

They grimaced.

"It's quite expensive."

If you'd like to talk about any of the points above, suggest a theme for next week's letter, or for the usual tips, leads, and conversation, feel free to send an e-mail to james.rundle@incisivemedia.com or call on +44207 316 9811. Next week's column will be from my colleagues, as I'm on annual leave.

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact info@waterstechnology.com or view our subscription options here: http://subscriptions.waterstechnology.com/subscribe

You are currently unable to copy this content. Please contact info@waterstechnology.com to find out more.

Systematic tools gain favor in fixed income

Automation is enabling systematic strategies in fixed income that were previously reserved for equities trading. The tech gap between the two may be closing, but differences remain.

You need to sign in to use this feature. If you don’t have a WatersTechnology account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here