[Beowulf] Collocation in Iceland?

Vincent Diepeveen diep at xs4all.nl
Wed Sep 5 11:40:31 PDT 2012


It's from 2010 and it's about arbitrage with a few assumptions.

His assumption is that from 1 location you need to reach more than 2  
exchanges.
This assumption is wrong.

To make money on the exchange in 99% of cases i'd guess all you need to
compare 2 different instruments, so that requirest 2 exchanges and  
based upon
that you can make money.

Using more than 2 means you'll lose money of course.

If i draw a straight line from worlds largest interesting exchange  
which is Chicago
and i draw a line from there to London and from London to mainland  
Europe,

Then Iceland is not in the middle of that, but about 2000 kilometers  
north of that line
through the ocean.

Furthermore it's questionable why one would need a line through a  
volcanic island,
i foresee that this would be a bad investment.

Now the next question this report raises, besides that he used wrong  
numbers to do his math,
starting with latencies; you can see that even a simple software  
program that IBM released had
already back in 2010 latencies from not much over 10 microseconds, so  
the math done by this
professor is a factor 50 off in his assumption; whereas some FPGA  
solutions already offered
it under 2 microseconds by then had he googled a tad.

So his assumption to build "in the middle" a datacenter is too much  
of a delay.

You want a direct cable connection from your computer at 1 exchange  
directly
to your computer at the datacenter of that other exchange.

Having a datacenter in Iceland is gonna cost lots of cash, say a  
million or 100,
and it's going to lose from the existing connections that are there  
which aren't
losing that 3000+ kilometers of distance nor get slowed down by yet  
another datacenter
in the middle.

The next mistake is the number of exchanges he assumes. He seems to  
not distinguish between
total uninteresting spots on the planet, such as New York, versus CME  
(Chicago)
which is rather interesting being worlds largest exchange where you  
can trade in say for
example futures.

Knowing there isn't too many exchanges on this planet and most are  
rather tiny or just for a particular
market, you'll end up with no more than half a dozen interesting  
exchanges to lay a cable to.

So He could have done a way easier research.

In fact we can also throw away some other exchanges. Say entire China  
we can skip.
They don't pay out much of a dividend in China. Probably ends up in  
their own pockets as
well as party members that are in power nearby that factory.

Based upon dividend payout, China is a very tiny nation.
So there is nothing to speculate about.

I see around a 7 big red dots in his report nearby China whereas only  
Japan is interesting.

Now i realize that this makes his math seem like useless as for his  
math he needs many datapoints
and assumptions. In fact he assumed there is 52 interesting exchanges.

Now maybe somewhere in future some interesting things will start to  
happen in Africa,
but we'll have to wait for that for now.

If such exchange in Africa gets interesting in the future a direct  
line from Chicago to Africa you do not
want to lay through a vulcanic island that in the first place had a  
majority of its population vote to steal cash
from saving accounts from The Netherlands and UK . (google for icesave)

It's not a properly run nation.

Kind Regards,
Vincent

On Sep 5, 2012, at 7:47 PM, Prentice Bisbal wrote:

>
> On 09/05/2012 10:23 AM, Joe Landman wrote:
>> On 09/05/2012 09:25 AM, Prentice Bisbal wrote:
>>> On 09/05/2012 12:38 AM, Joe Landman wrote:
>>>> On 09/05/2012 12:28 AM, Lux, Jim (337C) wrote:
>>>>
>>>>> But as long as we're talking quarrys and such, what about the  
>>>>> scheme of
>>>>> building a big pit to fill full of ice during the winter, and  
>>>>> melting it
>>>>> during the summer. (assuming you are in a
>>>>> less-than-wonderful-un-California-like climate where this would  
>>>>> work.)
>>>> I bet a good business case could be made for something like  
>>>> this.  The
>>>> west coast of Michigan is rich in having good wind availability  
>>>> and an
>>>> excellent (and visible from space) reservoir of very cold water  
>>>> that
>>>> could be used in heat exchangers to cool data centers.  Use the  
>>>> cool air
>>>> during the cooler months, and the cold water during the warmer  
>>>> months.
>>>>
>>>> Given the proximity to Chicago, I'd think it would make for a good
>>>> business to house LOTS of servers in a naturally well air  
>>>> conditioned
>>>> environment.    I can't imagine that the AC bill in Secaucus and
>>>> elsewhere in New Jersey is very pleasant.  Could ring the Great  
>>>> Lakes
>>>> with these.
>>>>
>>>> A shame I don't have the capital to do this.  Bet we could sell  
>>>> out the
>>>> space.  Without using oil, hydrogen/helium, or LN2.
>>> I don't think that would be a wise investment, since Iceland is  
>>> poised
>>> to become the colocation capital, due to it's cold temperatures and
>>> cheap, environmentally friendly geothermal energy. They're also  
>>> looking
>>> to increase the network bandwidth between Iceland and Europe and  
>>> North
>>> America. I brought this up on this list a couple of months ago:
>>>
>>> http://www.beowulf.org/pipermail/beowulf/2012-April/029613.html
>>>
>> Yeah ... but geography/location and other things conspire against  
>> this
>> becoming anything more than a niche player.
>>
>> In HFT latency matters.  Really ... seriously ... matters.  36
>> milliseconds to get to the DC?  Won't fly.  The folks we work with  
>> are
>> looking to shave microseconds off everything.  Anything and  
>> everything
>> is on the table for this.
>>
>> Extending the distance to your DC by a few thousand miles to take
>> advantage of cheaper power while losing out on the fastest  
>> trading?  Not
>> gonna happen.
> Changing subject since this is really a new thread.
>
> Joe,
>
> Did you read the second link in my post from the archives?  Here it as
> again, since I think I didn't copy-and-paste the URL correctly in my
> original post (the beginning of the URL was missing)
>
> http://dspace.mit.edu/handle/1721.1/62859
>
> There's a lot of HFT that involves trading between different  
> markets. I
> think this mostly the currencies/commodities. So someone might buy
> something on one exchange in NYC, and then sell it again on an  
> exchange
> in London or Tokyo.
>
> Because the HFT guys have reduced latency so much, the speed of  
> light is
> now a significant factor in their trades. In the paper above, an MIT
> Math professor calculated the best location to minimize latency  
> between
> different markets around the world. A couple of the locations he
> calculated are in Iceland. There's a few dots in Iceland, too.
>
> As someone who lives in the NYC metro area and can see firsthand the
> kind of money these HFT guys are throwing around, I would not be
> surprised to see data centers opening up in Greenland or Iceland,
> provided the networking is there. Heck, they're rich enough to put the
> networking there.
>
> --
> Prentice
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org sponsored by Penguin  
> Computing
> To change your subscription (digest mode or unsubscribe) visit  
> http://www.beowulf.org/mailman/listinfo/beowulf



More information about the Beowulf mailing list