Comments on: Getting the most recent record http://www.brentozar.com/archive/2009/04/getting-the-most-recent-record/ Your technology pain-relief experts. Wed, 08 Feb 2012 16:37:59 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: Brent Ozar http://www.brentozar.com/archive/2009/04/getting-the-most-recent-record/comment-page-1/#comment-10136 Brent Ozar Wed, 19 Aug 2009 15:30:25 +0000 http://www.brentozar.com/?p=3241#comment-10136 You're welcome! Glad I could help! You’re welcome! Glad I could help!

]]> By: Nancy D http://www.brentozar.com/archive/2009/04/getting-the-most-recent-record/comment-page-1/#comment-10134 Nancy D Wed, 19 Aug 2009 15:29:12 +0000 http://www.brentozar.com/?p=3241#comment-10134 Brent - Your original solution was so simple and it worked! I spent a huge chunk of yesterday trying to get it right and just couldn't. Found this site with google and am now bookmarking it. Thank you!! Brent –
Your original solution was so simple and it worked! I spent a huge chunk of yesterday trying to get it right and just couldn’t. Found this site with google and am now bookmarking it. Thank you!!

]]> By: Brent Ozar http://www.brentozar.com/archive/2009/04/getting-the-most-recent-record/comment-page-1/#comment-10083 Brent Ozar Thu, 13 Aug 2009 00:20:56 +0000 http://www.brentozar.com/?p=3241#comment-10083 Gerg - no, that's not a bug. If you only need 3 fields out of 20, you should only ask for those 3. Otherwise you're pulling unnecessary IO overhead by doing table scans when you might be able to take advantage of a clustering index instead, and you're pulling unnecessary network overhead by pushing those 20 fields (x every record) instead of just 3. Seriously, do yourself a favor and pick up Grant's book. It'll be an eye-opener for large-scale environments. I understand that doing SELECT * is easier for programmers, but if you need to scale to high volume environments, you need to watch every operation for performance tuning. Gerg – no, that’s not a bug. If you only need 3 fields out of 20, you should only ask for those 3. Otherwise you’re pulling unnecessary IO overhead by doing table scans when you might be able to take advantage of a clustering index instead, and you’re pulling unnecessary network overhead by pushing those 20 fields (x every record) instead of just 3.

Seriously, do yourself a favor and pick up Grant’s book. It’ll be an eye-opener for large-scale environments. I understand that doing SELECT * is easier for programmers, but if you need to scale to high volume environments, you need to watch every operation for performance tuning.

]]> By: Gerg http://www.brentozar.com/archive/2009/04/getting-the-most-recent-record/comment-page-1/#comment-10082 Gerg Wed, 12 Aug 2009 19:19:29 +0000 http://www.brentozar.com/?p=3241#comment-10082 That sucks. Have you reported bugs? You shouldn’t have to hard code ever field in every bit of your code making it hard to change schema without adjusting every layer. Don’t put up with bad tools, get them fixed! That sucks. Have you reported bugs?

You shouldn’t have to hard code ever field in every bit of your code making it hard to change schema without adjusting every layer. Don’t put up with bad tools, get them fixed!

]]>