October 8, 2009 at 5:23 am
i checked and the buffer cache is alway > 99% but the page life expectancy, during a heavy load (report), drops to 110 for a few minutes then ramps back up to about 750. the average is around 375 during the report time.
http://www.sqlmag.com/Article/ArticleID/43117/sql_server_43117.html
this article talks about a minimum target of 300 seconds then at the end talks of about bad average values. so is a drop to 110 for a few minutes necessarily that bad?
October 14, 2009 at 11:34 pm
Hi,
We have a/a/p setup of SQL Server 2005 EE x64 with SP3 on Windoes 2003 EE R2 x64 with SP2.
Node1 has 2 instances
Node2 has 4 instances
Node3 is passive.
Can we set "lock pages in memory" when having Multiple instances on each node?
If I set "lock pages in memory" , then how its going to help for multiple instances on a node?
Do we need to set this setting for Passive node too?
October 15, 2009 at 2:49 am
You can set 'Lock pages in memory' on all nodes. You must also work out how much memory to give each instance on each node so tat you do not over-commit memory.
You need to take all the normal things into account - the memory used by Windows, the memory you want eacj SQL instance to use, and the memory needed by non-SQL applications such as anti-virus, DB backup routines, etc. You aim is the same as for a non-clustered box: Make full use of the memory but avoid the need to do any paging.
Original author: https://github.com/SQL-FineBuild/Common/wiki/ 1-click install and best practice configuration of SQL Server 2019, 2017 2016, 2014, 2012, 2008 R2, 2008 and 2005.
When I give food to the poor they call me a saint. When I ask why they are poor they call me a communist - Archbishop Hélder Câmara
Viewing 3 posts - 16 through 17 (of 17 total)
You must be logged in to reply to this topic. Login to reply