Mon Oct 6 04:42:30 PDT 2008
- Next message: [Slony1-hackers] Slony-I Communications Costs
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
The Slony-I documentation states the following:
"This points to it being a bad idea to have the large communications
network resulting from the number of nodes being large. Up to a half dozen
nodes seems pretty reasonable; every time the number of nodes doubles, this
can be expected to quadruple communications overheads."
Consider a setup with dozens of very small nodes which aren't heavliy
frequented. Neither of these nodes is considered to be a forwarder and
stores only information which aren't updated frequently (so we consider a
small database between 20 - 30 MB). The Lagtime can be minutes (we don't
care wether 5 or 15 minutes). The network is stable.
I wonder wether the number of nodes can be significantly larger than the
half dozens the documentation mentions, especially if you don't need to use
Failover, MOVE SET and you are able to use high SYNC timeouts values. For
example, this can be configuration clusters of a large network where
configuration changes needs to be propagated through a large number of
nodes.
A customer is planning such a setup and they consider Slony-I exactly for
this purpose since they are impressed about its reliability. However,
because communications cost is quadratic, there can be significant network
traffic.
Opinions? Experiences?
--
Thanks
Bernd
- Next message: [Slony1-hackers] Slony-I Communications Costs
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Slony1-hackers mailing list