Data clustering is a very interesting topic and one which I have not analysed in great depth as yet. Usually it refers to methods of having a backup copy of your data, should a machine or hard drive fail.
The same way you can mirror a hard drive to provide redundancy, you can also mirror a whole machine. While this is more expensive, it offer great enhancements to securing your data against hardware failure.
Usually the term "Data Clustering" is used with "server farms", a stack of servers which are interconnected between them and can seamlessly take over any service, should any server fail, and all this is transparent to the users.
I am actually looking at implementing such a setup at work, but that won't be until later on this year.
With Windows 2000 (any possibly any advanced O/S), Data clustering can be defined a method of providing full redundancy for your data. This could mean from hard drive mirroring or raid 5 hard drive configuration, to server mirroring.
In a medium size windows 2000 network, installing at least two servers with active directory and mirrored drives (or even better, raid 5) means that you are safe should a hard drive that contains critical data fails, or even the system drive, as no Active dir. information is lost since all servers are equal.
I plan to introduce disk mirroring, duplexing, spanning e.t.c at sometime, but it wont be soon !
Hope I have helped, if I confused you, then its probably because I myself am confused .... its 2:16 am, and I just got back from work... [img]images/smiles/icon_smile.gif
I would also point you to [url=http://www.whatis.com,]www.whatis.com,[/url] its an excellent site for any IT term/definition.