Re: Dean's performance document

[Rodent of Unusual Size <co...@decus.org>]

    It just occurred to me that it might be worthwhile to recommend that
    documents that are subject to negociation be kept as small as
    possible, since they'll be hitting the wire repeatedly unlike cached
    ones.  It'll increase the hash tables for the caches, but so what?
    The consumable to be protected is usually the bandwidth..

    Is that a good idea or am I off base, O Performance Freak?

    #ken    P-)}

Re: Dean's performance document

[Dean Gaudet <dg...@arctic.org>]

HTTP/1.1 clients can cache negotiated documents ... but I'm not sure if
you're referring to the sort of "fake" negotiation that we can do when
using multiviews to serve a single document where the extension isn't
predetermined?  If so CacheNegotiatedDocs should be set in that case yup. 
Otherwise yeah negotiation sucks with 1.0 clients.  There's also
mod_expires.

Dean

On Sun, 14 Sep 1997, Rodent of Unusual Size wrote:

>     It just occurred to me that it might be worthwhile to recommend that
>     documents that are subject to negociation be kept as small as
>     possible, since they'll be hitting the wire repeatedly unlike cached
>     ones.  It'll increase the hash tables for the caches, but so what?
>     The consumable to be protected is usually the bandwidth..
> 
>     Is that a good idea or am I off base, O Performance Freak?
> 
>     #ken    P-)}
>