Transparent failover and restore?

[Brad Fitzpatrick]

I'm not even talking about using MySQL at all.  Just NDB cluster.  That's
the underlying daemons and protocol that MySQL cluster is built on.

The MySQL memory tables suck for about two dozen reasons.



On Sun, 19 Dec 2004, Kevin A. Burton wrote:

> Brad Fitzpatrick wrote:
>
> >On Sun, 19 Dec 2004, Kevin A. Burton wrote:
> >
> >
> >
> >>MySQL cluster is still a bit early ... It also doesn't provide disk
> >>based storage (all memory) so hence it can be very expensive to deploy.
> >>
> >>
> >>
> >
> >But memcached doesn't have disk-based storage.
> >
> >
> >
> I know... but memcached is an LRU cache. MySQL cluster is not.
>
> >At least with MySQL Cluster you can stop/start a node without losing any
> >data.
> >
> >It does back up the dataset it's storing in memory to disk, though.  And
> >if it hasn't committed in a while, it can get the data from another node.
> >
> >Anyway, I'll stop with the MySQL cluster discussions for now, at least
> >until I have any code or benchmarks.
> >
> >
> >
> Its interesting but its still really early. A more conservative approach
> would be to use the memory tables...
>
> Anyway...
>
> --
>
> Use Rojo (RSS/Atom aggregator).  Visit http://rojo.com. Ask me for an
> invite!  Also see irc.freenode.net #rojo if you want to chat.
>
> Rojo is Hiring! - http://www.rojonetworks.com/JobsAtRojo.html
>
> If you're interested in RSS, Weblogs, Social Networking, etc... then you
> should work for Rojo!  If you recommend someone and we hire them you'll
> get a free iPod!
>
> Kevin A. Burton, Location - San Francisco, CA
>        AIM/YIM - sfburtonator,  Web - http://peerfear.org/
> GPG fingerprint: 5FB2 F3E2 760E 70A8 6174 D393 E84D 8D04 99F1 4412
>
>