Viewing 15 posts - 16 through 30 (of 33 total)
Has this query been regularly as OK? or is it a new query?
Are the statistics up to date?
Have you checked sys.dm_exec_requests for more detailed information?
October 31, 2012 at 11:02 am
Focus on consistency - is the PLE consistently low?
Have you checked for missing indexes, updated statistics?
Try and establish why there is memory pressure.
Increasing physical memory will alleviate pressure , but...
October 30, 2012 at 9:44 am
Make sure all live transactions are stopped , if there is a transaction ongoing and writing to that log file you will not be able to delete the Log.
October 30, 2012 at 9:40 am
Convince your management of paying for a "first - line support" - someone who works out of hrs - can pick up all the monitoring , deal with failed backups,...
October 24, 2012 at 11:51 pm
Some considerations, although these are general guidelines and all implemenattions should be tested against real data sets . Firstly, don't forget to maintain statistics.
1) Check the WHERE clause , if...
October 23, 2012 at 11:39 pm
As long as you have the sqlcmd file installed on the server from which you're executing the code, use the SQLCMD method to log onto the relevant servers. Check SQL...
October 23, 2012 at 11:22 pm
What scripting framework are you using? Powershell, batch files with sqlcmd etc.
October 22, 2012 at 11:52 pm
First thing you must do when you start is ensure there is a Test Server\ laborator server. Start practising various recovery scenarios - and be fanatical about BACKUPS. I know...
October 22, 2012 at 12:26 pm
@bugg - Ask them details on RPO and RTO of database server. Also , enquire about any Service level agreements they may have with other groups. This will...
October 17, 2012 at 11:46 pm
Do they have a Security Policy? If not , maybe that is an opportunity for you to implement. You can also develop reports which audit the security .
October 15, 2012 at 12:29 pm
It depends on how large is the environment . How motivated you are to learn new skills. Focus on fundamentals , such as availability - incluing backup strategy ,...
October 15, 2012 at 12:25 pm
I agree server-side trace is a good approach. Another low-intrusive method is to tune the queries against the execution plans.
August 22, 2011 at 11:41 pm
For justifications:
1) high io_completion wait stats
2) high io write and io read -use sys.dm_io_virtual_file_stats, this provides useful data regarding IO statistics. An example...
August 6, 2011 at 8:03 am
Adding to comments already made.
1) Identify , whether it's a pre deployment or post deployment query. If pre- then it must be optimised, if post, then optimise if a problem....
July 26, 2011 at 8:11 am
Depending the amount of data changes , and on your STATISTCS management, some data distribution may have changed. As a first step , Force a UPDATE STATISTICS and ...
July 24, 2011 at 12:32 am
Viewing 15 posts - 16 through 30 (of 33 total)