SSWUGtv – Amazon Web Server and MS SQL Server
With Stephen Wynkoop
In this Week’s Technology Bit – What issues might you have moving to Amazon Web Services if you use SQL Server?
Watch the Show
Preparing for Deprecation
I started reading about the deprecated list by following a link from Jeff, sent in response to my editorial on sp_depends yesterday.
Jeff says:
One of the key takeaways I got from your comment about the replacement for sys_depends in SQL Server 2008 is to use four part names when doing a cross database join.
This feature is listed as being deprecated in a future version of SQL Server according to the following article: http://technet.microsoft.com/en-us/library/ms143729.aspx
I totally agree with Jeff’s assessment. Three and Four part identifiers will not be supported sometime in the future. The list of features being deprecated in SQL Server 2012 feels larger than past releases. That’s just an un-qualified gut reaction to the rather large list. As I look at the list it seems to be highly influenced by the SQL Azure platform.
I think this list impacted me in that there are a number of features for which I will need to start planning in order to continue upgrading to future releases. Most of them make sense, or are enforcing better practices. Some not so much.
So, how about you? When do you start planning to address deprecated features in SQL Server? Are there features in this list you find important to you? Get into the conversation by writing btaylor@sswug.org.
Q & A
Naveen asks, “Is there any way to defrag tables of SQL 2008? I do know that we can do that for indexes, but could not find anything related to tables.”
This is a frequent topic here and in many blogs. We covered this topic in detail a few months ago, so you’ll find a great deal of insight in our newsletter archives. In the meantime, the key to your question is that only indexes may be defragmented. Only tables, based on a clustered index, may be defragmented by defragmenting the clustered index.
Microsoft seems to like a clustered index based on a sequential number. There are some options to this solution, but they come from much older advice. The issue revolves around maintaining the BTree in which the index is maintained. So, the simplest solution, without a lot of detailed learning is to use an IDENTITY column as a clustered index key. If you want to go further, look more deeply into the way a clustered index is maintained and utilized.
Cheers,
Ben
$$SWYNK$$
Featured Article(s)
Learning SQL Server Analysis Services: Part 1- Introduction
This is first part of the series dedicated to learning SQL Server Analysis Services. This article will serve as an introduction to the technology and acquaint you with the basic terms of the field.
Featured White Paper(s)
Go Beyond SCOM Monitoring with Foglight for SQL Server
read more)
Featured Script
sp_MY_sqlmaint_rndx
This procedure will run DBCC DBREINDEX and sp_recompile against all user-defined tables within whatever database it is run in… (read more)