- From: David W. Morris <dwm@shell.portal.com>
- Date: Thu, 19 Oct 1995 00:39:15 -0700 (PDT)
- To: http working group <http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com>
On Mon, 16 Oct 1995, Jeffrey Mogul wrote: > show more locality than client-server interactions I observed, since > the client has fewer proxies to choose from than it has servers. I know of no clients which allow specification of more than 1 proxy for the http protcol (it might be a good idea...) so I suspect that to the extent the client uses a proxy, it will be a single proxy. > client A and client B are sharing the same connection, and both > start retrievals at the same time?" In HTTP-NG, we might I would expect that the proxy might have multiple connections to the same server when needed to avoid this difficulty. It would seem reasonable for the proxy to keep the connection allocated to the same client until the client switched destinations or the client-proxy connection was terminated. In that case, repeated transfer of access control, etc. might be reduced. We would need a reset signal however to allow the proxy to signal a change of client. > So, even though one of our papers suggested that combining > multiple clients onto one proxy-server connection would be > more efficient, I'd now recommend against implementing this > in HTTP 1.x. I don't have a conceptual problem with such a restriction, I started this sub-topic because in looking at implementing a special purpose proxy I got concerned and thought we needed to deal with the issue either by working out the design or 'ruling' out the function. Dave Morris
Received on Thursday, 19 October 1995 00:44:33 UTC