![]() | |
![]() |
| | Thread Tools | Display Modes |
#11
| |||
| |||
|
|
Hi All, Our cluster server has three nodes (active/active/active), total Physical Memory on each node is 32 GB. In this cluster, I installed three instances (SQL Server 2005 with SP2 Windows 64 bit), every instance located on the specific node. Current configuration as follow: Instance1 min server memory and max server memory equal to 10 GB, Instance2 min server memory and max server memory equal to 10 GB, and Instance3 min server memory and max server memory equal to 8 GB. So if all three instances fail over to one node, the OS still has 4 GB free memory. But if there are setting on own node (normal status), the rest of around 22 GB memory seems use less. I try to set up min server memory for every instance 10 GB, max server memory 20 GB, but what happy after fail over two instances on the same node? Let’s say, Instance1 used 20 GB (max), Instance 2 fail over to same node as Instance 1 and increase memory usage to 12 GB. Is this node will frozen? I want SQL using as much as memory they can. What’s the best way to configure (memory) in this situation? Thanks for your input! Chen |
![]() |
| Thread Tools | |
| Display Modes | |
| |