December 16, 2010 at 10:02 pm
Comments posted to this topic are about the item Under Appreciated Features
December 16, 2010 at 10:30 pm
One thing that immediately comes to mind is the Upgrade Advisor. For ISVs planning for compatibility with an upcoming version of SQL Server, the first thing they should do (I use the word should because technology-aware organizations do, upcoming ISVs don't even know such a utility exists) is to run the Upgrade Advisor. Major issues come up right away and per the 80/20 rule, 80% of your issues are now in front of you awaiting resolution within 20% of the time.
Next is the power of connection strings. Quite often I have seen just the basic 5 parameter connection string (Provider, Server, DB, User, Password) being used in ISV products. Connection strings are much more than this, and their power needs to be demonstrated.
Have a great Friday, and a rocking week-end!
Thanks & Regards,
Nakul Vachhrajani.
http://nakulvachhrajani.com
Follow me on
Twitter: @sqltwins
December 16, 2010 at 10:33 pm
I would have to say Service Broker as one. There is a lot one can do with it, and quite a bit of the necessay infrastructure is already built. The problem is that you still have a lot that needs to been done manually and it isn't all that intuitive.
Wizards would help, especially if you get to look at the SQL it builds for you instead of just doing it for you (like Database Mirroring - Would benefit from being able to see the scripts that wizard builds and the order they are run).
This is one way some people learn, show me what you do at the basic level, and let me learn how to do the more complex stuff by building on that.
December 17, 2010 at 1:46 am
Was I alone in getting excited about 'filtered indexes'? (Is that what they're called? An index with a where clause)
The reason I got into a state is that it enabled me to relatively simply accomplish something that previously had obliged me to jump through hoops.
That is to place a unique index on a nullable column. To address the requirement of a column which is nullable but which, when it is populated, must be unique for the table.
David McKinney.
December 17, 2010 at 3:16 am
I think SQL Management Studio is actually under-appreciated. I was just having to do some work on a client's SQL 2000 install, and I'd forgotten what a pain it is to have to run a separate query analyzer application to check things out! (Yes, I know *real men* use SQLCMD anyway, but you know what I mean :-)).
December 17, 2010 at 4:38 am
By Toutatis, I hate to be the first naysayer, but SSMS has to be the biggest pile of steaming crap ever to be the excuse for a dev env. Using it is like trying to stride through molasses. I am constantly aware of the productivity it is costing me. Are two tools such an imposition, given that they work much better, and about three times faster ?
I don't use the RedGate tools, but this must be about the spot for a plug for them 😉
On the positive side, I'd like to cite the improvements in the optimiser. Problem queries (that shouldn't be) in SQL2000 run sweetly in SQL2005. It's like oxygen - you only notice it when it's not there.
Now if only I could get SQL2005 to run with the old tools ....
Ben
December 17, 2010 at 4:54 am
I'm torn between Service Broker and missing indexes being shown in the graphical plan. The latter, whether you know how to tune or not, is a real time saver. No, you can't blindly follow all the suggestions, but I bet I create the suggested indexes 95% of the time.
December 17, 2010 at 5:06 am
I think a cool feature of sql server 2005 and above that you don't always hear a lot about is the recursive query with the CTE. We use this a lot. In our software product we have a Bill Of Material (BOM) structure. This BOM structure starts with a parent Item Nbr and then lists which children Items make up the parent. Some of our clients have BOM structures that go quite deep. Anyway, a recursive query with using a CTE has worked really nicely to return the whole bom structure for a parent item quickly and easily.
Ben
December 17, 2010 at 5:47 am
To a certain extent any new feature, because we don't trust Microsoft to not pull the rug out from under us.
You mentioned "DTS lead to SSIS"
More like Microsoft threw DTS under the bus and brought out SSIS.
I think the most under appreciated feature was the ability to create system functions.
Clearly it was under appreciated since they yanked it.
It was really nice having a business rule in one place that could be called from any database without a three part name. A bonus was being able to create and test a new version in a test database and then updating the system version.
Microsoft's arrogance and hubris in deciding what they think is best for us (don't get me started on Office 2007) and their total lack of commitment to backwards compatibility leads many of us to not trust them.
December 17, 2010 at 6:25 am
I would say catalouge views and DMVs added in 2005. Those really changed the ways that DBAs get meta-data and troubleshoot problems. But DMVs are still not widely adopted. They have lots of potential.
In programing, window functions like ROW_NUMBER() is a real help.
December 17, 2010 at 6:43 am
Sometimes the simplest things are the most useful - my favorite feature of SQL 2005 : the ability to filter in the object explorer. We have several databases with thousands of Queries and being able to jump to (or close to depending on how specific the filter) the right query without scrolling through an endless list saved me loads of time.
December 17, 2010 at 6:54 am
I don't think you can overlook the usefulness of bcp and/or bulk insert. They're frequently overlooked in favor of SSIS. Many times SSIS is overkill for loading/producing simple files.
I have a love/hate thing going with intellisence.
On the wish list is the ability to bundle procedures and functions into a class (similar to oracle packages).
--Paul Hunter
December 17, 2010 at 6:58 am
I have to say I am a big fan of Tuning Advisor. Some of the databases I work with are an absolute nightmare for performance and the nature of the company means we always need our data fast, (it's disheartening for users and reflects bad on me when things are slow). This is where I find Tuning advisor to come up trumps! 🙂
December 17, 2010 at 7:05 am
+1 for filtering in the Object Explorer of SSMS
+1 for partition functions
+1 for execution plans (both graphical and xml)
New ones - not mentioned so far.
XQUERY - Processing xml has improved a lot, it still has some way to go, but is much better.
FOR XML PATH - just for the ability to concatenate strings
Query Optimiser - That is one major bit of kit - without it we would all be wallowing in SQL 7 hell.
SSC - not strictly a part of SQL server - but certainly integral to my use of SS.
MM
select geometry::STGeomFromWKB(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ecember 17, 2010 at 7:09 am
It might not sound like much but being able to highlight a string and press F1 to get help on the selected text is absolutely awesome in SSMS- it's not something you notice until you have to use MySQLs workbench or query browser and this functionality doesn't exist.
All those times when the order of the arguments in an arcane function are a bit hazy, or when you need to know how it handles nulls and other little bits of trivia and you get it at the press of a button! An instantaneous answer on it is truly invaluable. On MySQL, it's open up the net, navigate to their webpage, navigate to the right section or use their hideously slow search engine and keep scrolling down until you find what you're looking for, 5 minutes later and you only just remember what you were looking for in the first place.
Viewing 15 posts - 1 through 15 (of 67 total)
You must be logged in to reply to this topic. Login to reply