Just as a point of interest:
> Now, does akamai simply cache stuff out of the blue? No idea. It
> seems that people may need to have a biz relationship with akamai
> first. *shrug* My point wasn't to provide a concrete example, but to
> point out that a caching network *could* create some scaling
> benefits for SVN repositories.
Akamai's caching is provider-driven (unlike more traditional caching,
where the client selects the proxy server). The general way a page is
"akamized" is that the IMG tags on the page contain URLs like
"http://___.akamai.net/___/www.apple.com/promo/powerbook/images/
pdf_icon.32x32.gif" where ___s are replaced by a bunch of internal
Akamai gook. The "real" URL at the end tells the caching network
where it can get the actual data if it doesn't have it.
I think if the real server HTTP redirected someone to one of those
funny Akamai URLs at the beginning, everything would work (for
read-only operation), since I assume the libsvn_ra_dav tree walk is
done by appending to the parent directory's URL. But I'm not familiar
enough with the code to know for sure.
(And yes, you have to have a business relationship with Akamai before
you can legitimately use their caching network as a provider, of
course.)
Received on Sat Oct 21 14:36:12 2006