The Final Frontier: 500 Microseconds Between Wall Street and Chicago
A couple of months ago, there was a big scandal over the fact that someone apparently learned about a Fed decision sooner than they should have. It takes seven milliseconds for a signal to travel from Washington DC to Chicago over a fiber optic cable, but a couple of big orders were placed on the Chicago exchange a mere couple of milliseconds after the Fed announcement. Shazam!
But if an advantage of a few milliseconds is so important, why bother with fiber optic cables? Why not mount repeaters on blimps or something, and then relay wireless signals? At the speed of light, it would only take about four milliseconds from DC to Chicago.
I suppose I should have guessed, but naturally someone is doing this:
Ari Rubenstein, a "Star Trek" fan who counts physics as a hobby....heads Strike Technologies, a New York company that's part of a budding cottage industry racing to build networks of ultra-fast microwave radio transmitters linking the world's financial hubs.
....Strike, whose ranks include academics as well as former U.S. and Israeli military engineers, hoisted a 6-foot white dish on a tower rising 280 feet above the Nasdaq Stock Market's data center in Carteret, N.J., just outside New York City.
Through a series of microwave towers, the dish beams market data 734 miles to the Chicago Mercantile Exchange's computer warehouse in Aurora, Ill., in 4.13 milliseconds, or about 95% of the theoretical speed of light, according to the company.
Remember that Keynes thing about goosing the economy by burying money in landfills and letting people dig it up? In terms of social utility, this strikes me as about the same thing. It's hard to imagine millions of dollars being spent more uselessly. Even gold-plated toilet seats probably have more value to society than this.
In any case, I still think my idea for a neutrino communications network that transmits directly through the earth is a better bet. Sure, you'd need a million gallons of chlorine or heavy water or something to act as the detector, but that seems pretty trivial in order to save another 500 microseconds. Who's going to be the first to do this?