Has anyone tried how this monster http://webaccelerator.google.com/index.html does affect the ability to retrieve wikipedia pages?
Mathias
I can't test it since it's Windows-only, but seems some sort of caching scheme. Thus, for logged-in users it won't make any difference, since pages are customized and you need to hit wikimedia's servers.
For non logged-in users it might be useful.
Alfio
On Thu, 5 May 2005, Mathias Schindler wrote:
Has anyone tried how this monster http://webaccelerator.google.com/index.html does affect the ability to retrieve wikipedia pages?
Mathias
Wikitech-l mailing list Wikitech-l@wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikitech-l
Mathias Schindler wrote:
Has anyone tried how this monster http://webaccelerator.google.com/index.html does affect the ability to retrieve wikipedia pages?
Mathias
Hoi, I think the idea behind it is brilliant. But consider what happens when Google starts HOSTING Wikimedia content. Their webaccelerator combined with WMF cache functionality will result in even faster service again. The trick for Google is that their service doing the web acceleration must be local so it follows that hosting the WMF projects will be one of the most interesting bits of research for Google. The size of our data is not too big, currently the amount of our traffic is basically as much as we can provide server wise.. With Google hosting us, they could throw addiditional hw at the problem as demand for our product increases.
Combining hosting and acceleration will give you a hosting platform that will be stellar and commercially hard to beat. Because they have the servers and they can do the on demand computing as the market requires it..
Thanks, GerardM
wikitech-l@lists.wikimedia.org