https://issues.apache.org/SpamAssassin/show_bug.cgi?id=5845





--- Comment #4 from Theo Van Dinter <[EMAIL PROTECTED]>  2008-03-15 14:14:41 
PST ---
Woo, mapreduce! :)

The issue with all this stuff, of course, is that it's written to deal with
processing a large dataset with processing nodes in close proximity -- ie: a
LAN w/ low latency, packet loss, high bandwidth, and probably no need for
things like SSL, etc.

Our model is very not that. ;)

We don't have a single dataset to run over, we have several individual (and
unlikely to be shared) datasets.  Without sharing the data, we end up
immediately reducing our setup to individual processing clusters, where the
norm is to only have 1 machine (or at most a small handful of machines).

So there's really two sets of issues here:

a) We want <distributed clusters> to use a standard set of rules and send in
the results.  It doesn't really matter if it's net/weekly/per-commit, it's the
same request essentially, though there are different ways to implement handling
them since two are periodic and long-running, and the other is randomly timed
and short-running.

b) In each of the distributed clusters, we possibly want to farm out the
incoming request to multiple machines for faster processing.

Hadoop could (and our current client-server system does) handle issue b, but it
really doesn't help for a, imo, which is what I think you're trying to target
here.


-- 
Configure bugmail: 
https://issues.apache.org/SpamAssassin/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

Reply via email to