[Probably OT]: Data replication
a.einstein at unforgettable.com
Tue Oct 3 12:23:37 EDT 2000
I've read this mailing list for some weeks and I recognized I probably misunderstood the meaning of cluster. This mail is to better understand what this word means and to ask something I really and absolutely need.
In my wrong old idea, a cluster was a group of computer doing something "together". This something could be anything: parallel computing, load balancing for various kind of servers --i.e. web server-- and, this is what I needed, data replication.
I've seen a demo in which some Microsoft techies had 2 Digital AlphaServers with Windows NT installed. These servers where used as database servers: they had their separate array of disks and where used in some form of load balancing.
After some minutes of --boring-- demo, they unplugged the power chord of one Digital without shutting down the system. On the other Digital's monitor a popup said that the other system crashed so it had to answer all the queries of the client. And everything went on without any kind of problem.
This is what I'd need to do: I need to have a couple of machines --2 or 3 would be enough-- with real-time synchronization of stored data. Load balancing isn't that important --I mainly use them as smb file server-- but data replication is absolutely necessary. If a machine crashes or hangs or any piece of hardware get out of oder, I need one of the other machines do its job without anything to do for the users. And shared data must be coherent on all the machines. Since I don't have very heavy network traffic, it's ok to setup one machine as master and the other as slave, if this can help me/us to setup this configuration.
I hope I was clear enough for you to suggest me some software package to use and is Beowulf can do this kind of data replication.
I hope this wasn't too off-topic for this list.
Gianni "BugBarbeq" Rondinini.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Beowulf