Re: Implementing a new caching algorithm.

Henrik Nordstrom wrote:
> On Fri, 11 Mar 2005, Nick Lewycky wrote:
>
>> is a branch called "collapsed forwarding"[4], but it currently only
>> applies to accelerator setups. It could perhaps be adapted.
>
> It appliest equal to all setups. It was designed for accelerator setups
> as there is where this is most needed, but is not limited to acceleration.

That's good news.

> The drawback is that latency is increased significantly on frequently
> used non-cacheable content as the second client needs to wait for the
> first reply to arrive before Squid knows the reply can not be cached.

That's an interesting problem. So for a non-cacheable resource, n hits
from the client must correspond to n hits to the server? Otherwise you
could send the result to both clients, yes?

What's keeping it from being landed on the Squid trunk, perhaps with a
configuration option? I've been considering rebasing my own branch off
of it, I just haven't gotten around to trying it out. I was assuming it
was unfinished.

Nick
Received on Fri Mar 11 2005 - 20:13:23 MST