Tag Archives: computing

Location, location, location

Why would anyone in their right mind consider building a server farm in deepest darkest Siberia, or the middle of the Indian Ocean? Possibly because the intersection of geography and information flow means such locations would give you a slight yet crucial edge in the high-stakes imaginary-money game of high-frequency trading [via SlashDot]:

The insight of the MIT researchers, Alexander Wissner-Gross and Cameron Freer, is that some automated traders–or at the very least, their server farms–will be best positioned in-between certain exchanges. Since some trading strategies capitalize on price fluctuations between separate exchanges in different parts of the world, the optimally located server will receive information from those exchanges at precisely the same moment, gaining that millisecond advantage over the competitor. In some cases that pefect location is the midpoint between the two exchanges, but not always–it depends on whether the exchanges’ prices move at the same speed or not.

Wissner-Gross and Freer rounded up the locations and price-speeds on the 52 largest global exchanges, and plotted a map of the ideal locations for traders who would want to be perfectly positioned between any given pair. The map, which appears today in an article in the journal Physical Review E, dictates that some traders’ servers will be ideally positioned in central Africa, others in the remotest forests of Canada, others in the middle of the Indian Ocean, and still others in Siberia. This all assumes, of course, a proper infrastructure in place–in the short term, Freer tells Fast Company, it might make more sense to approximate these locations, rather than invest in installing a server farm underneath the ocean.

Brilliant… yet another way for compulsive gamblers to squeeze more profits out of the aether (not to mention shades of Ian McDonald’s Dervish House – which, if you haven’t read it yet, should be added to your stack of pending reads with immediate effect). But according to New Scientist, this might actually represent the last possible way to grasp advantage in the automated trading system:

“This shows that the technological arms race to extract every penny from high-frequency mechanical arbitrage will soon reach its ultimate limits,” says physicist and hedge-fund manager Jean-Philippe Bouchaud, based in Paris. “Maybe the buzz around high-frequency trading will then calm down.”

We can live in hope, I guess.

Quantum computing for dummies

Heard people talking about quantum computing, but not really sure you understand what they mean? Well, you’re far from alone (as the late great Richard Feynman once said, “anyone who claims to understand quantum physics doesn’t understand quantum physics”), but why let that stop you from trying to get a layman’s grasp of the basic ideas?

That, one assumes, is the spirit in which this brief introduction to quantum computing at Silicon.com has been written [via SlashDot]… though I’m in no position to comment on how accurate or useful it is. Input from passing physicists is, as always, more than welcome. 🙂

Hang on, what’s quantum entanglement when it’s at home?

I was afraid you were going to ask. Quantum entanglement is the point where scientists typically abandon all hope of being understood because the thing being described really does defy the classical logic we’re used to.

An object is said to become quantumly entangled when its state cannot be described without also referring to the state of another object or objects, because they have become intrinsically linked, or correlated.

No physical link is required however – entanglement can occur between objects that are separated in space, even miles apart – prompting Albert Einstein to famously dub it “spooky action at a distance”.

The correlation between entangled objects might mean that if the spin state of two electrons is entangled, their spin states will be opposites – one will be up, one down. Entangled photons could also share opposing polarisation of their waveforms – one being horizontal, the other vertical, say. This shared state means that a change applied to one entangled object is instantly reflected by its correlated fellows – hence the massive parallel potential of a quantum computer.

Accuracy aside, what’s interesting to me is seeing this sort of bluffer’s guide in a venue like Silicon.com, which is more of a business organ than a tech one. Prepping the Valley VCs for upcoming investment decisions, perhaps?

Fractal levels of simulated reality, forsooth!

I’m sure I ran a story similar to this a while back, but I’m damned if I can find it in the Futurismic archives, so I’m gonna mention it anyway: it’s the one about the folk building logic-based processors within the virtual spaces of computer games, the latest example being the insanely popular (and rather lucrative) Minecraft. Find blocks of material with the right in-game properties, chain ’em together, and hey presto, you’ve got a simulated arithmetic processor made of non-existent lumps of an entirely fictional substance. Whole lotta meta, right there.

I think the reason I love these stories is because of the extrapolatory end-point: the implication is that given simulated spaces of sufficient size and complexity (and sufficient player-hours, or clever macros to obviate the need for such), one could build a computing device within that simulation which was itself capable of running a simulation within which another computing device could be simulated. Sort of like Nick Bostrom rewriting Lavie Tidhar’s “In Pacmandu”… it’s simulated turtles all the way down! Now, where’s the door back to my origin reality, please?

Probabilistic processing: the analogue computer waits in the wings

Digital processing has the advantage of versatility – the utter ubiquity of computer technology is a testament to that. But digital logic has to use lots of bits to represent large ranges of values; perhaps some applications – spam filtering, for instance, or pattern analysis – would run better and faster on a system that allowed for analogue values “in the raw”, so to speak?

Lyric’s innovation is to use analogue signals instead of digital ones, to allow probabilities to be encoded directly as voltages. Their probability gates represent zero probability as 0 V, and certainty as VDD. But unlike digital logic, for which these are the only options, Lyric’s technology allows probabilities between 0 and 1 to use voltages between 0 and VDD. Each probabilistic bit (“pbit”) stores not an exact value, but rather, the probability that the value is 1. The technology allows a resolution of about 8 bits; that is, they can discriminate between about 28 = 256 different values (different probabilities) between 0 and VDD.

By creating circuits that can operate directly on probabilities, much of the extra complexity of digital circuits can be eliminated. Probabilistic processors can perform useful computations with just a handful of pbits, with a drastic reduction in the number of transistors and circuit complexity as a result.

This could so easily be an excerpt from a Rudy Rucker story… or a Neal Stephenson novel, for that matter.

Ditch the keyboard and mouse, go with Skinput

Although my aesthetic tastes tend toward the more retro versions of cyberpunk style (born in the final few years of Gen X, can’t help it), I’m still very seduced by the sheer pragmatic awesome of using your body as an input device for your portable hardware [via SlashDot].

Need to turn down the volume on your PMP? No problem; just jab a finger at your left forearm.

[… the] Skinput prototype is a system that monitors acoustic signals on your arm to translate gestures and taps into input commands. Just by touching different points on your limb you can tell your portable device to change volume, answer a call, or turn itself off. Even better, Harrison can couple Skinput with a pico projector so that you can see a graphic interface on your arm and use the acoustic signals to control it.

Projector, pah. A proper cyberpunk would get the controls tattooed on there instead. 🙂