Collections: Crossing intervals computed gaps

Question

Post reply

Collections: Crossing intervals computed gaps

Bogdan SAHLEAN

SSC Veteran

Points: 219
More actions
March 23, 2020 at 12:00 am

#3729660

Comments posted to this topic are about the item Collections: Crossing intervals computed gaps

Viewing 4 posts - 1 through 3 (of 3 total)

You must be logged in to reply to this topic. Login to reply

Jeff Moden SSC Guru Points: 1003851 More actions · Answer 1

How is the performance for a large data set?

--Jeff Moden

RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
First step towards the paradigm shift of writing Set Based code:
________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.

Change is inevitable... Change for the better is not.

Helpful Links:
How to post code problems
How to Post Performance Problems
Create a Tally Function (fnTally)

Bogdan SAHLEAN SSC Veteran Points: 219 More actions · Answer 2

First, thank you for your interest in my article. I have done performance testing just for this article

https://www.sqlservercentral.com/articles/collections

comparing proposed solution vs table vars using single thread tests and also concurrent tests (3 concurrent connections) and comparing CPU, logical reads, elapsed time and the results where positives favoring the proposed solution (maybe because of less stress on tempdb), biggest difference being record for logical reads. For this solution, I'm expecting to get (at least) not the last results just because spatial objects are SQLCLR data types.

Jeff Moden SSC Guru Points: 1003851 More actions · Answer 3

Just to be sure, I appreciate anyone willing to step up and write articles to share their knowledge with other people especially when they come up with such a different method.

Getting to the point, though... I expect no performance gains just because something is an SQLCLR type behind the scenes. In fact, because of things like the FORMAT function, I've grown to being extremely skeptical when it comes to expectations of many built-in SQLCLR backed functions.

With that being said, I was looking for some form of performance testing (or the results of such testing) in both this article and the one you cited above. I'm just not seeing anything that might resemble a test harness. To be sure, I normally test things with a million rows. In this case, I would probably test with a million intervals that contain at least 50,000 gaps of 1 to 1000. Then I'd do the same thing with some of the more commonly known methods.

Who knows? You might be right. Maybe the SQLCLR code behind the scenes was written much better than things like the FORMAT, et al, function.

--Jeff Moden

RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
First step towards the paradigm shift of writing Set Based code:
________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.

Change is inevitable... Change for the better is not.

Helpful Links:
How to post code problems
How to Post Performance Problems
Create a Tally Function (fnTally)