Undiscovered Country of HFT: FPGA JIT Ethernet Packet Assembly
michaelmalak writes "In a technique that reminds me of the just-in-time torpedo engineering of Star Trek VI: The Undiscovered Country, a company called Argon Design has "developed a high performance trading system" that puts an FPGA — and FPGA-based trading algorithms — right in the Ethernet switch. And it isn't just to cut down on switch/computer latency — they actually start assembling and sending out the start of an Ethernet packet simultaneously with receiving and decoding incoming price quotation Ethernet packets, and decide on the fly what to put in the outgoing buy/sell Ethernet packet. They call these techniques 'inline parsing' and 'pre-emption.'"
This is the madness of high-speed trading...
No sig. Move along - nothing to see here.
These people could be used to develop more efficient hardware for everyone to use, or fix medical conditions, rather than make rich traders even richer at the expense of another economic collapse. It seems wrong that our economy prioritizes high frequency trading so much.
Things like high frequency trading make me want to vom. Essentially, all they're doing is shuffling money around, taking advantage of an outdated system, and increasing risk for the entire world.
It'd be great to see this kind of innovation in something that actually is useful and valuable - not for creating an incremental improvement on a corrupt system.
Fucking parasites (and their toolmakers).
No, this is arbitrage. Taking advantage of price differences within or between markets.
now I wouldn't have a problem with it if it was accessible to everyone, for example if anyone could buy machine time from vm's that were all given the information at the same time(artificially arranged, wouldn't work otherwise!) at the stock exchange.
Sure you would. You might not think so, but suppose that the exchange set up a perfectly equitable system in which identically configured were made available to every firm and provided with identical market feeds all perfectly synchronized so that no single trading VM has any advantage over any other.
I would give that system about three hours of run time before you discover that:
Just look at the tricks that players in the game are already using, and ask yourself how changing the rules is going to stop them.
I have a friend who works in the defense industry but interviewed with HFT firms around 3 years ago. My friend also had this idea, and discussed it with me since I am close to the industry.
If an industry outsider like my friend had this idea within a week or two of merely interviewing for jobs, it is a good bet many others had already conceived it and even gotten it working before that.
Yes, and when I was 8 I had the idea for a nuclear-powered photon drive for a rocket ship. Unfortunately translating the concept from my crayon drawing of a Saturn 5 with light-bulbes instead of exhaust nozzles to a working prototype proved to be impractical.
Ideas are cheap. Implementations are what matters. Particularly for something like this where hardware and software both need to be developed and speed and reliability are paramount. Having a basic idea of the algorithm is a far cry from a devise that works reliably enough and faster enough to be an improvement over established competing systems.
We need to stop rewarding folks for high-speed trading. It basically steals money from the folks genuinely invested in the companies whose stock is traded and adds no value to the system at all.
Moderating "-1, Disagree" is simple censorship. Have the guts to post your opinion.