[dns-operations] co-operative cacheing

Tony Finch dot at dotat.at
Thu May 8 10:49:08 UTC 2008


On Wed, 7 May 2008, Paul Vixie wrote:
>
> one could theoretically offer a "forward-first, recursion-undesired,
> short-short-timeouts" config option, to opportunistally fetch an answer
> from a "nearby" server, and then head out to authority-land otherwise.

I was thinking of using a local broadcast to perform the cluster lookaside
(though you'd need a way of deciding who gets to reply), perhaps in
parallel with the normal query so that it doesn't affect latency.

> there is an ideology out there called "distributed hash tables" which i
> know opendns.com uses to solve exactly this problem.

I've read one or two papers along these lines. They generally seem to be
focussed on wide-area services, whereas my scenario is a fairly tight
cluster. However a DHT would be useful for a more advanced version of my
idea, where the cache is partitioned instead of replicated across
the cluster.

> what i've been thinking about for a couple years now is a way to send each
> upstream answer received by a recursive nameserver, to other recursive
> nameservers who are thought of as "administratively nearby", so that all
> servers would have a chance to cache things that any server was asked.

That could be another way of achieving my goal.

Thanks for your reply.

Tony.
-- 
f.anthony.n.finch  <dot at dotat.at>  http://dotat.at/
ROCKALL: SOUTHEAST 5 OR 6 VEERING WEST 3 OR 4. MODERATE. SHOWERS. MODERATE OR
GOOD, OCCASIONALLY POOR.



More information about the dns-operations mailing list