[MUD-Dev] World Persistence, flat files v/s DB v/s ??

[Matt Chatterley]

On Sun, 22 Mar 1998, Vadim Tkachenko wrote:
> Joel Dillon wrote:
> > On Sat, 21 Mar 1998, Ben Greear wrote:
> > 
> > >
> > > I'm contemplating a space based game, where once started, it will
> > > never again resemble it's starting state (unless the starting world
> > > image is saved of course.)  It will be written in Java.
> > >
> >   I've been thinking along these lines (a non-resetable mud) except
> > I'm using C++.
> > 
> > > The game will need to update it's persistant storage very often
> > > to make this feasible.
> > 
> >   How often is very often? Wouldn't a backup and save to disk every
> > half-hour or so be ok?
> 
> Why don't you think about some abstract number instead of half-hour?
> What about 'as soon as needed', and make it incremental, so only changed
> data is being queued for update, and update is being performed in a
> background low-priority thread?

A valid notion. My mud is being set up so that rather than the (standard
for LPmuds as far as I know) system of rebooting once per period of time
(usually 24-48 hours), it will reboot at any time within a period of 6
hours when there are no user connections open to the server.

Obvious advantage: Players are never inconvenienced by reboots.

Obvious disadvantage: What if there is never such a gap? Well,
statistically speaking, if it checks every 6 hours, it'll check 28 times a
week (and these times will reflect the period at which it last rebooted,
so it will hit the same time again). Should it go a period of y days with
no reboot, it will reduce the frequency of checks. At some time, the mud
will be empty. :P

I was briefly toying with the idea of rigging a system so that users were
not disconnected during reboots. Unsure on this currently.

[Snip rest]

--
Regards,
	-Matt Chatterley
Spod: http://user.super.net.uk/~neddy/spod/spod.html