Viewing 15 posts - 13,531 through 13,545 (of 49,552 total)
I don't know a reference offhand, but I can tell you absolutely, any read of any page, regardless of the isolation level, will take a shared page latch for the...
April 11, 2013 at 5:04 am
Repeating the exact same sentence doesn't explain anything.
Do you want indexes that have more than 50% logical fragmentation? If so, see my previous reply. If not, then please explain what...
April 11, 2013 at 3:39 am
Nolock != nolatch
Nolock or the read uncommitted isolation level simply mean that a query does not take shared locks while reading. All other locks and latches are taken and held...
April 11, 2013 at 3:35 am
What do you mean by 'more than 50% of the database'?
If you want indexes that have a logical fragmentation higher than 50%, just query sys.dm_db_index_physical_stats and filter on the logical_fragmentation_in_percent...
April 11, 2013 at 3:32 am
Ellen-477471 (4/10/2013)
Would it be a percentage of the existing...
April 10, 2013 at 3:30 pm
456789psw (4/10/2013)
But doing more reading, its seems just because they are named subquery does not mean a temp table(worktable) is not created in the background.
True, but the same goes for...
April 10, 2013 at 10:43 am
456789psw (4/10/2013)
If the CTE is using recursion then its uses a spool operator therefore tempdb is used....therefore not much performance benefit vs just creating a temp table
How exactly would...
April 10, 2013 at 9:57 am
456789psw (4/10/2013)
CTE's seem to be using tempdb just as using a temp table or table variable.
CTEs are nothing more than a named subquery. They are not temp tables or table...
April 10, 2013 at 8:49 am
The very first thing I will suggest is patch that instance. You're running RTM of SQL 2005 which had some nasty problems in it. Patch up to SP4 ASAP.
April 10, 2013 at 2:21 am
ericjorg (4/9/2013)
GilaMonster (4/9/2013)
Good way to destroy a database...http://sqlinthewild.co.za/index.php/2009/06/09/deleting-the-transaction-log/
Yes, that is exactly right. In my case, this was a test database and the logfile was created in a different version...
April 9, 2013 at 1:08 pm
ericjorg (4/9/2013)
I had this issue on Sql 2012 and fixed it this way:1) Stop SQL
2) Delete the recovering database's log file
3) Start SQL
Good way to destroy a database...
http://sqlinthewild.co.za/index.php/2009/06/09/deleting-the-transaction-log/
April 9, 2013 at 12:49 pm
SQLSACT (4/9/2013)
GilaMonster (4/9/2013)
Are you just querying large tables or modifying them? If just querying, then you won't see any lazy writer activity as it has nothing that needs writing.
I'm just...
April 9, 2013 at 8:53 am
Are you just querying large tables or modifying them? If just querying, then you won't see any lazy writer activity as it has nothing that needs writing.
April 9, 2013 at 8:22 am
SQLKnowItAll (4/8/2013)
It is pretty clear that with auto_update_statistics enabled, there is no reason whatsoever to recompile after the statistics update.
There's never any need to recompile after a stats update no...
April 9, 2013 at 2:03 am
Viewing 15 posts - 13,531 through 13,545 (of 49,552 total)