[squid-users] Inconsistent accessing of the cache, craigslist.org images, wacky stuff.

[Amos Jeffries]

On 29/10/2015 3:02 p.m., Jester Purtteman wrote:
> but my bigger question is:  if I setup a parent
> proxy that ONLY grabs the big updates down on my big-fast-cheap
> connection, then set my little-slow-expensive-connection up to pull
> from that connection, would that have a higher chance of success?
> Since the proxy on the slow system is requesting the same object, I'm
> wondering if that may work out better.  Not sure that will have the
> desired effect, but I'm going to try it out, I'll let you know how
> that works out.

I dont quite grok that sorry. Can you diagram what you are thinking?

With a front-end proxy you would start to see revalidation requests
happening between the proxies. Due to many origin servers still sending
out new content even if it has not changed, this setup can result in
small bandwidth savings just by existing. The main gain is helping to
optimize the traffic bandwidth and reducing TCP connection count over
long connections like satellite links, where the target is optimal g the
bandwidth behaviour (reducing it is just part of that).

If the frontend has a bigger cache than the backend you will see churn
and extra bandwidth consumption as repeats get served from the frontend
cache. But the origin traffic upstream of it will stay low. This is good
if the internal links are fast and upstream is slow. Like most LAN
situations. It is usually best to have a cache on the client side of a
choke point (slow connection).

Amos