January 11, 2008 at 11:57 am
I was asked a question today if in SQL Server 2005, if the actual database can be a active/active setup. Or is it just a failover cluster? Say I have a 3 node cluster, with two active and one passive, can both the active nodes be active on the database? This would allow me to use resources on both boxes at the same time for all the databases using the SQL cluster.
Hope that makes sense.
JW
January 11, 2008 at 12:08 pm
As of right now - the way to have 2 actives and one passive is to have 2 instances being serviced by 3 machines. Meaning - the two nodes aren't active on the same instance - they're active on different instances (which presumably would "split" the one instance into 2 halves, so as to spread the activity).
the drives aren't "shared" either. Meaning - at any given time - only one instance has access to a given data volume.
Depending on what you're trying to do - there might be different methods (other than clustering) to get multiple servers "sharing the work". One common thing to do would be to use replication or mirroring to move a read-only copy onto another server for reporting purposes. One not so common thing would be to partition a database across multiple database servers in what is called a Federated database setup: however - each of the individual "node" servers only houses part of the data.
----------------------------------------------------------------------------------
Your lack of planning does not constitute an emergency on my part...unless you're my manager...or a director and above...or a really loud-spoken end-user..All right - what was my emergency again?
January 11, 2008 at 1:06 pm
Matt Miller (1/11/2008)
As of right now - the way to have 2 actives and one passive is to have 2 instances being serviced by 3 machines. Meaning - the two nodes aren't active on the same instance - they're active on different instances (which presumably would "split" the one instance into 2 halves, so as to spread the activity).the drives aren't "shared" either. Meaning - at any given time - only one instance has access to a given data volume.
Depending on what you're trying to do - there might be different methods (other than clustering) to get multiple servers "sharing the work". One common thing to do would be to use replication or mirroring to move a read-only copy onto another server for reporting purposes. One not so common thing would be to partition a database across multiple database servers in what is called a Federated database setup: however - each of the individual "node" servers only houses part of the data.
Appreciate the follow up.
I was looking for this information, because we are trying to find out what is the best way for us to setup a SQL cluster. As of now, we have roughly 12 applications (medium to small) that will need a SQL backend piece. This could easily grow as time goes on.
Knowing that, we were trying to find out what type of hardware we would need for this to work. That is where I am a little fuzzy on how SQL clustering works.
As it stands now, I need to find out solutions for this cluster.
Any further suggestions or things to consider?
Appreciate it.
Jason
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply