Hello Chuck;
Another thing that's impressed me with OpenBase is the clustering
failover. I doubt many people have used this feature.
Any SQL written to any database in the cluster is automatically
propagated to other OpenBase databases in that cluster within five
seconds without any primary key collisions.
Do you know how OpenBase handles that? This is usually quite the
problem with clustering and replication. Does it need to have a
GUID as the primary key, or does it work with integers?
As far as I can remember (Scott, you can correct me if I am wrong) it
uses a ticker-tape style transaction log -- effectively "log
shipping". In this manner, transactions are written to a log and then
re-played on the other members of the cluster. To avoid primary key
conflicts, one employs the OB primary key generation plugin. By using
this, one member of the cluster issues primary keys 10,20,30 and the
next 11,21,31,41 etc... Off the top of my head, I can't remember how
conflicts between logs get resolved -- again, perhaps somebody from
Openbase can remind us.
cheers.
___
Andrew Lindesay
www.lindesay.co.nz
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Webobjects-dev mailing list (Webobjects-dev@lists.apple.com)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
