Quote:
|
Originally Posted by bendsley
Windows Server 2003 provides several methods of improving system- or server-level fault tolerance by using a few of the services included in the Enterprise and Datacenter platforms.
Also, if you build machines that are there strictly for replication, you should just be able to change your router to point to the new machine, assuming it's replicating data on the fly.
This link might help you some too.
I do have one question for you though, what RAID are you using? We run a quad xeon machine with MS SQL on it in RAID 1 over 6 harddrives in three arrays. We have found that for SQL, RAID 1 is actually faster than the RAID 5 alternative. Everything else we have uses the RAID 5. |
Awesome link, thanks! Some of our database servers don't even have raid, and I believe our others have raid 0+1. But also, as I said, sometimes a single database server overloaded and would be nice to let the excess load spill over to another, less active machine. Can this only be done thru clustering?
A buddy made a suggestion of having a load balancer split up the work between, say 4 machines, and have them all replicate. Should 1 machine go down, the load balancer would see this and not use that machine until it sees that it is up again. Would this work?