We have a Windows 2003 Enterprise MSCS cluster with three physical nodes (8G
memory each) and two virtual instances. Our default configuration is to host
each virtual instance on a physical node and host the clustering services on
the third (spare) node. I have seen the article that discusses AWE and
failover clustering:
http://msdn.microsoft.com/en-us/libr...0).aspx?ppud=4
in which it states that:
If you are using SQL Server 2005 failover clustering and AWE memory, you
must ensure that the sum of the max server memory settings for all the
instances is less than the least physical memory available on any of the
servers in the failover cluster. If the failover node has less physical
memory than the original node, the instances of SQL Server 2005 may fail to
start or may start with less memory than they had on the original node.
My question is does this apply if we have a 'spare' node. Since we have one
more physical node than virtual it seems we only need to set the max server
memory if we anticipate loosing two physical nodes. If we lost two physical
nodes then both instances may not fit in a single physical node but I feel it
is unlikely that we would lose two physical nodes and would rather allow each
virtual instance to use the full physical memory rather than constraining
each instance to only use half the available memory.
Thanks for your comments,
Michael
Linear Mode
