[time-nuts] nanoseconds in the news
bownes at gmail.com
Fri Feb 17 00:50:33 UTC 2012
There was a system in NJ with over 3000 ps2s in a supercomputer config.
The financial industry's idea of high cost is a bit different than most.
The price of your real time ticker feed from the exchanges is directly proportional to the associated network latency due to speed of light.
On Feb 16, 2012, at 18:30, Tom Knox <actast at hotmail.com> wrote:
> EndRun set up the system next door to the NYSE used by GS. I think EndRun has a white paper on their site describing the ultimate in insider trading systems. It has been the subject of a number of articles and a whistle blowers case. Some believe it was responsible for a 1000 point "glitch" several years ago.
> Thomas Knox
>> From: namichie at gmail.com
>> Date: Fri, 17 Feb 2012 09:20:07 +1100
>> To: time-nuts at febo.com
>> Subject: Re: [time-nuts] nanoseconds in the news
>> Obviously no one is thinking big enough.
>> A hyper-super computer on either side of the Atlantic could run a model
>> of each of the stock markets which could be synchronised by frequent
>> data transfer.
>> The learning power of these models would be very great, and they
>> could deliver a real-time
>> estimate of the other markets that would be hundreds of milliseconds
>> better that neutrinos.
>> The long term veracity of these computers could not be doubted, the
>> noise in short terms
>> would be continually reduced.
>> A system like GPS could be used to synchronise them so they could
>> give nanosecond response.
>> Would that not be great for huge spikes!
>> Neville Michie
>> On 17/02/2012, at 9:07 AM, Peter Monta wrote:
>>>> Time is money...
>>> I wonder if long-distance neutrino links might be attractive to the
>>> financial community. The SNRs are currently way too low, but with
>>> aggressive engineering, a link through the Earth would shave off many,
>>> many milliseconds (even at not greater than the speed of light :-) ).
>>> The cost would be very high though.
>>> time-nuts mailing list -- time-nuts at febo.com
>>> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/
>>> and follow the instructions there.
>> time-nuts mailing list -- time-nuts at febo.com
>> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
More information about the time-nuts