[Spread-users] Number of machines in a configuration

[J C Lawrence]

On Sat, 31 Jul 2004 10:23:13 -0400 
Ryan Caudy <rcaudy at gmail.com> wrote:

> Well, the limit isn't all that hard in terms of what the code can
> support, in general.  

Excellent.

> You should just need to change a few things (I'm not sure of all of
> them, but MAX_PROCS_RING is probably the most important one).
> However, the real limitation is the increase in token latency this
> might bring.  You may have to tweak the membership.c timeouts in order
> to allow your network to work with a few hundred members.  Also, the
> latency to deliver many message types will be increased proportionally
> to the number of daemons in your ring.

Aye, that I can deal with.  For the application in question even quite
slow service is acceptable assuming that the other guarantees are
maintain.

Thanks.

-- 
J C Lawrence
---------(*)                Satan, oscillate my metallic sonatas.
claw at kanga.nu               He lived as a devil, eh?
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.