Viewing 15 posts - 5,626 through 5,640 (of 5,841 total)
I advise my clients that when they create a new production database they size it to hold the estimated total data volume (including indexes) for 12-18 months of data plus...
August 23, 2007 at 1:57 pm
Prior to disk defrag, you should shut down all sql server services. Just for giggles, did you size all the files initially to some appropriately large value with an...
August 23, 2007 at 12:44 pm
Mark, you didn't answer my second question about whether the order of the rows matters. If I move these rows into another table for processing, can I order them...
August 23, 2007 at 12:27 pm
is there an identity key on this table, or is what you show the entire table definition? also, does the initial order of the rows need to remain intact?...
August 23, 2007 at 9:06 am
A few possibilities:
1) The amount of data in the tables has grown such that different query plans become 'least costly'. This can result in DRASTICALLY different query plans and...
August 23, 2007 at 8:42 am
I would certainly try recompile first. There are probably very few situations where this won't routinely result in the optimal query. Plus having a query plan developed for...
August 22, 2007 at 12:43 pm
I wasn't aware that you could "disable parameter sniffing". Do you mean use WITH RECOMPILE on the sprocs perhaps?
Also, the user stated that the exact same data...
August 22, 2007 at 11:40 am
1) You didn't explicitly mention if the SQL Server was the EXACT same version (sp and hotfixes).
2) While doing your development, you should have made your test platform as similar...
August 22, 2007 at 9:26 am
select
distinct 'CONV 1' as Conveyour,
info as Error,
(select top 1 substring(timecreated, 0, 7) from log b where a.info = b.info order by timecreated asc) as Date,
(select count(*)...
August 22, 2007 at 9:01 am
If by analyzing profiles he can identify the actual calls that are causing performance problems then the OP can (hopefully) easily grep through the source code to find where the...
August 20, 2007 at 12:28 pm
Be VERY careful with where you put your "WHERE" clause components when doing JOINs. With OUTER joins you can get vastly different results depending on whether the AND is...
August 17, 2007 at 7:57 am
The most important thing about the 'old' syntax is that it will not be supported in the future per Microsoft. The *= and =* is already unsupported in SQL...
August 16, 2007 at 7:33 am
Management Studio is a GUI for connecting to database servers. There is a link to it in the Start Menu folder that was created when you installed SQL Server....
August 15, 2007 at 7:32 am
I have some client's with "unfortunate datastructures" that do require cursor processing. I also use them occassionally when looping through system objects for one or another reason. So...
August 14, 2007 at 9:38 am
1) clustered indexes generally help with BETWEEN queries.
2) they do have a down side in that the clustering key is carried as the pointer for every non-clustered index on the...
August 14, 2007 at 9:29 am
Viewing 15 posts - 5,626 through 5,640 (of 5,841 total)