On Tuesday 10 May 2005 12:44, Patrick Okui wrote:
> On Tuesday 10 May 2005 01:26 pm, JFL wrote:
> > Hi.
> > Does anyone here have info on how this works
> > http://webaccelerator.google.com/
>
> More like squid on steriods - the idea is google has a
> cache of most web based content *and* they've been
> building servers in all sorts of locations. This
> together with the weird way they can redirect you to
> the "closest"...

Source IP-based global load balancing. Akamai sort of do 
the same thing, not sure if they are doing it for Google 
though.

> server AND they presumably have loads of 
> bandwidth makes them a theoretically good cache.
>
> I *think* that in addition it maintains a local cache
> on your machine of sites you visit, but I'm not sure
> about that.

From what I've seen with these *Web Accelerators*, they 
don't cache, per se, but just open multiple sessions to 
the same remote host, or open distributed sessions to 
multiple hosts for the same content, and in effect, use 
up all available frequencies in a pipe.

The browser continues to cache the objects as part of its 
inherent design. 

Mark.

>
> Has its pluses, but some think it should be
> redesigned.
>
> http://radar.oreilly.com/archives/2005/05/google_web_a
>cce_1.html
>
>
> --
> patrick
> _______________________________________________
> LUG mailing list
> [email protected]
> http://kym.net/mailman/listinfo/lug
> %LUG is generously hosted by INFOCOM
> http://www.infocom.co.ug/
_______________________________________________
LUG mailing list
[email protected]
http://kym.net/mailman/listinfo/lug
%LUG is generously hosted by INFOCOM http://www.infocom.co.ug/

Reply via email to