Today I received the unhappy news that my current position at Microsoft has been eliminated, effective this Friday.  This was not entirely unexpected, but I find that the anticipation of a sad day does precious little to ease its arrival.

It’s been a great eleven years-plus working with our customers and the SQL Community.  I’ve been part of some truly exciting efforts, including the first-ever large-scale use of SQL Server’s XML capabilities for Metropolis/Agent Desktop (we started coding on pre-Alpha SQL Server 2005 bits).  Just this year, I’ve contributed my architecture and coding skills to an important new internal system which  is changing the way my former team manages its IP, and to the new SQL Census component of SQLRAP, which is changing the way T-SQL Coding IP is delivered to MSFT’s SQLRAP customers.

I think I’m proudest of my involvement in the MCA and MCM Database programs.  What was then the SQL Server Center of Excellence – Mark Pohto, Tim Wolff, Steve Harrell, Greg Moss, and myself – wore jack-of-all-trades hats through the first four rotations.  While the program has evolved considerably since then, I will always believe that those first four rotations were very special.

It has been an honor and a privilege working with such great people at such a great company.  I hope we’ll be able to revisit the experience sometime soon.

If you’d like to stay in touch, please use my personal email, ward@wardpond.net, or visit my website, www.wardpond.net.  I'll continue to maintain my personal blog at pondsponderings.wordpress.com, and I'll soon be starting a new technical blog at pondonsql.wordpress.com

In the meantime..  this space is closed..  at least for now.

Peace..

     -wp 

-wp

this copyrighted material was originally posted at http://blogs.technet.com/wardpond. 

the author and his employer are pleased to provide this content for you at that site, and via rss, free of charge and without advertising.

the author welcomes and appreciates links to and citations of his work.  however, if you are viewing the full text of this article at any other website, be aware that its author does not endorse and is not compensated by any advertising or access fees you may be subjected to outside the original web and rss sites

It’s a very disturbing trend in one’s life when so much of one’s writing begins with the three letters which this entry does.

It’s an even more disturbing event when the subject is one’s father.

It’s my sad task to report that my father, Thomas Alexander Pond, passed away early Sunday.  He was eighty-five.

All, all that you dream
Comes to shine in silver lining
And clouds, clouds change the scene
Rain starts washing all these cautions
Right into your life, make you realize
Just what is true, what else can I do
Just follow the rule
Keep your eyes on the road that's ahead of you

My father was a gifted educator whose work touched thousands of people.  You can read more here about his contributions to Rutgers University, one of the institutions he served.

We’ll have more to say soon; in the meantime, I’m with my sister, sharing memories and making plans for Dad’s memorial.

I’ve said this before many times, but it bears repeating: if it’s cosmically possible, please call your parents.

-wp

this copyrighted material was originally posted at http://blogs.technet.com/wardpond. 

the author and his employer are pleased to provide this content for you at that site, and via rss, free of charge and without advertising.

the author welcomes and appreciates links to and citations of his work.  however, if you are viewing the full text of this article at any other website, be aware that its author does not endorse and is not compensated by any advertising or access fees you may be subjected to outside the original web and rss sites

Sad, sad news today from the worlds of music and friendship: Richie Hayward – founding member, drummer, writer, and singer in the seminal American rock and roll band Little Feat – lost his year-long battle with liver cancer today.  He was only 64.

I first became familiar with Richie and Little Feat when I was in the tenth grade.  My neighbor and surrogate older brother, Erich Swartz, returned home for Thanksgiving and brought with him a copy of Time Loves a Hero.  To say I was hooked was an understatement – within a week I owned their entire discography, and I’ve been a passionate FeatFan ever since.  Thirty-five years, in round numbers, including the lean years in the eighties when there was no new Little Feat music.

Like all great American stories, Little Feat had an incredible second (and third..  and fourth..) act, reuniting in 1988 to great acclaim.  By then, I had moved to Los Angeles and hooked up with the Maltose Falcons Brews Band.  Our bass player at the time, Dennis Fink, knew the band!  He lived near several of its members in beautiful Topanga Canyon, which is also home the Will Geer Theatricum Botanicum, an amazing open-air theater in the woods.  Dennis came to rehearsal one night and told us that Little Feat was playing a benefit for the theater (“Feats for Seats”), as several of their family members had served on its Board of Directors over the years.  He also relayed an invitation from the band – if we schlepped their gear in and out of the theater, we could join them for dinner and watch the show.

My response now would be the same as it was then.. what’s the question?

That night, I met my musical heroes, and I was welcomed into the amazing Little Feat family.

In the mid-‘90s, when Little Feat took over their own promotional work and used fan volunteers to great effect, I was the regional FeatFan coordinator for Oklahoma, Mississippi, and Arkansas.  When I moved to Seattle in 1999 to work for Microsoft, I sent an email to the community informing them why I was resigning my post.  I shortly got an email from Paul Barrerre, asking, “hey, can you get software?”

Paul being the generous man he is, he offered to trade me back-stage passes for the software.  Pretty soon, he stopped asking for software and started asking how many passes I needed.  These folks who I’d regarded as titans of rock and roll since I was fifteen became, in my late thirties, my friends.

Every time they come to town, Gale and I go to the show and hang with the band, and every time, we spend a little more time there.  I recall particularly a gig they played at the Triple Door in Seattle.  Richie was in fine fettle that night, and he probably spent half an hour with us after the show.  Gale took particular interest in Richie’s tour book, a minute-to-minute guide that the road manager assembles for each member of the band when they’re on the road.  Richie went over it with her in great detail, and confessed that he’d never seen anyone outside of the band quite so interested in a tour book.

The best musicians show you a clear path to their heart through their playing.  The expression on Richie’s face in the photograph to the right says it all.  When Richie would come off-stage after a particularly epic show, his sweaty face would split into an ear-to-ear grin as he said, “man, that was fun.”

It sure was, Richie, for you and for us, for your entire career.  Who’d’ve thought that a five-word want ad placed by Lowell George in 1969 (“Drummer wanted.  Must be freaky.”) would lead to such an incredible forty-year-plus ride?

My thoughts and prayers go out tonight to Richie’s incredible wife, Shauna, his son, Severin, and the rest of the Hayward and Little Feat families.

The world is slightly less syncopated today than it was yesterday.  A great talent, a beloved husband and father, and a cherished colleague has left us.  Rest in peace, Richie, and thanks to you and your family for all the years you spent on the proud highway, bringing joy through your music.

Now that you’re up in heaven with Lowell, there must be one hell of a jam going on.

-wp

this copyrighted material was originally posted at http://blogs.technet.com/wardpond. 

the author and his employer are pleased to provide this content for you at that site, and via rss, free of charge and without advertising.

the author welcomes and appreciates links to and citations of his work.  however, if you are viewing the full text of this article at any other website, be aware that its author does not endorse and is not compensated by any advertising or access fees you may be subjected to outside the original web and rss sites

• http://WardPond.net, my personal website, and
• http://pondsponderings.wordpress.com, my personal blog

While you’ve been in no danger of being deluged by personal traffic here lately, there have been times in this blog’s history where that’s not been the case.

It’s my hope that the differentiation which the new blog facilitates will better serve those who seek both types of insight.

Please let me know if you have any questions or concerns, and, as always, many thanks for your continued interest.

-wp

this copyrighted material was originally posted at http://blogs.technet.com/wardpond. 

the author and his employer are pleased to provide this content for you at that site, and via rss, free of charge and without advertising.

the author welcomes and appreciates links to and citations of his work.  however, if you are viewing the full text of this article at any other website, be aware that its author does not endorse and is not compensated by any advertising or access fees you may be subjected to outside the original web and rss sites

So, as energy, inspiration, time, and circumstances permit, this humble little corner of the InterNets will continue to provide a forum for discussion of safe and sane care and feeding of SQL Server databases.

Thanks for being a part of it.

-wp

this copyrighted material was originally posted at http://blogs.technet.com/wardpond. 

the author and his employer are pleased to provide this content for you at that site, and via rss, free of charge and without advertising.

the author welcomes and appreciates links to and citations of his work.  however, if you are viewing the full text of this article at any other website, be aware that its author does not endorse and is not compensated by any advertising or access fees you may be subjected to outside the original web and rss sites

Jens has a great post regarding the internal behavior of various forms of sp_-named programmability objects.  Some SQLRAP-related research led me to take a different tack, and I wanted to share the results of that research here as well.

So here are five things about sp_ to be aware of.  I’ve verified all of these behaviors on SQL Server 2000 and above (through SQL Server 2008 R2).

1. If we load onto a user database a programmability object named identically to one of the MSFT-supplied sp_ procedures (“sp_who”, “sp_lock”, “sp_monitor”, etc.), that code will never be executed.  The code in master will always run, even if we issue a three-part call to your user-created object.
   
2. If we load such a programmability object into model, we’re essentially replicating issue 1. above in every subsequently created database.
   
3. If we load our own sp_ programmability object (with a name which doesn’t match that of a MSFT-supplied object) in a user database and another identically named object in master, things get even more complex.
   
   In the user database where the programmability objects resides, a one-part call will execute the copy in the local database -- the opposite of the behavior for MSFT-supplied sp_ objects discussed in item 1. above.
   
   In other user databases, a one-part call will execute the copy in master – identical to the behavior for MSFT-supplied sp_ objects described in item 1.
   
   Confused yet?
   
   What if the code in these two objects is different?  Then a one-part call will behave differently in different database contexts.
   
   What if the object name exists in multiple user databases, each with different code under the name?  The problem multiplies.
   
   
4. By this same logic, I trust it’s clear that we wouldn’t want to load multiple copies of our hypothetical sp_mycode into any user databases without loading a copy into master.  Other than an affection for the convention, there’s no real reason to do this, and confusion can result if we subsequently load a new or copied sp_mysproc into master.
   
   
5. The only possible permutation which remains, then, is a single instantiation of sp_mysproc in master.  Lots of people do this and survive just fine, but given the foregoing, this approach strikes me as an unnecessary potential source of instability.
   
   If the objects are instead housed in a dedicated database, there are several ways to access them via a one-part call.  Use synonyms where available (SQL Server 2005+); in SQL Server 2000, load like-named objects on each user database which simply execute the three-part call to the object in the dedicated database.  We could even load such objects in model to ensure that our utility objects are accessible from newly created databases.
   

So -- surprisingly enough on a Microsoft-hosted SQL Server blog -- my recommendation is that you leave the sp_ object names to Microsoft, and use one of the workarounds discussed above to implement a naming convention which behaves consistently across your instance.

The headaches you save will likely be your own.

Does your shop have any rules regarding use of sp_?  Do you have any stories to tell -- good or bad -- about using it with user-created objects?

Contact me if you have any questions or comments, or if you’d like simple repro scripts which demonstrate these concepts.

-wp

this copyrighted material was originally posted at http://blogs.technet.com/wardpond. 

the author and his employer are pleased to provide this content for you at that site, and via rss, free of charge and without advertising.

the author welcomes and appreciates links to and citations of his work.  however, if you are viewing the full text of this article at any other website, be aware that its author does not endorse and is not compensated by any advertising or access fees you may be subjected to outside the original web and rss sites

Please find time soon, if you’re able, to call your mother and thank her for your life.  Such precious opportunities grow rarer by the day.  Please don’t squander as many as I did.

Happy birthday, Mom.  I love you, and I miss you every day.

-wp

this copyrighted material was originally posted at http://blogs.technet.com/wardpond. 

the author and his employer are pleased to provide this content for you at that site, and via rss, free of charge and without advertising.

the author welcomes and appreciates links to and citations of his work.  however, if you are viewing the full text of this article at any other website, be aware that its author does not endorse and is not compensated by any advertising or access fees you may be subjected to outside the original web and rss sites

Last summer, in the posts Database Programming: What I’ve Learned About SQL Server 2008 (with a little on SQL Server 2005 thrown in) and Database Programming: Why I Like MERGE, I commented that I’d been able to build a MERGE statement which included XML access within a CTE to perform our basic data maintenance in a new system we were building.

Well, when we got around to performance testing this code, performance was pretty miserable, with logical reads on a worktable in the tens of millions when processing eight thousand rows.

My testing indicated that at least part of the issue related to the use of XML access within a CTE; the entire XML document was being rebuilt and reparsed with each iteration through the CTE. 

The solution in this case was to refactor the query by placing the results of the XML access into a temp table, and accessing that structure in the MERGE statement.  This refactoring also removed the CTE, and resulted in a two-orders-of-magnitude performance improvement.

So..  I still like MERGE, but CTEs using XML access?  Not so much.

-wp

this copyrighted material was originally posted at http://blogs.technet.com/wardpond. 

the author and his employer are pleased to provide this content for you at that site, and via rss, free of charge and without advertising.

the author welcomes and appreciates links to and citations of his work.  however, if you are viewing the full text of this article at any other website, be aware that its author does not endorse and is not compensated by any advertising or access fees you may be subjected to outside the original web and rss sites