[j-nsp] Qfabric

[Saku Ytti]

On (2011-02-25 11:31 -0800), Joel Jaeggli wrote:
 
> network infrastructure is a very small portion of the capital outlay for
> one of these operations so refreshing equipment to take advantage of
> small opportunities is easy to justify...
> 
> That said, once you've gone cut-through, shed all the stateful devices
> in the path and reduced the rtt from 40km to 150meters you are correct
> that focus on other portions of the system is going to dominate.

Thanks Joel, this and especially the rationale that it doesn't make dimple
in their revenue stream, so as long as it doesn't make things worse, it's
worth the investment.

I'm still slightly ignorant about the true source of market data. I would
assume that if you want to be high frequency trader, you need to locate
your servers in same building as stock exchange for every exchange you want
to trade in. This probably doesn't scale. as not everyone can fit their
trading cluster in same building.
I'm drifting rather far from j-nsp topic, but might be interesting topic
for nango preso, addressing the problem and solution(s) end (ticker origin)
to end (pc making trading decision)

To contribute something remotely relevant to thread this may look familiar
http://www.broadcom.com/products/Switching/Carrier-and-Service-Provider/BCM56840-Series

-- 
  ++ytti