RE: SQL Date Math

["Wilt, Charles" <WiltC@xxxxxxxxxx>]

Michael,

Those are interesting results.

I'm a little surprised at the difference between having the code inline vs. in a UDF with an index.

I'm thinking you might be seeing the effects of some of the data still being in memory.  Consider that
in your first you say that CvtDateToNumISO took only 39 seconds vs. 120 seconds.

There's a recommend procedure for clearings memory pools and what not between benchmark runs, but I
don't know it off the top of my head.  Sure somebody else on the list does or has a link to the info.

As a quick and dirty check, trying running the tests in reverse order:
Using function code with index - ???
Using CvtDateToNumISO function with index - ???

Charles

> -----Original Message-----
> From: midrange-l-bounces@xxxxxxxxxxxx 
> [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Michael Ryan
> Sent: Wednesday, November 14, 2007 8:30 AM
> To: Midrange Systems Technical Discussion
> Subject: Re: SQL Date Math
>
> That was interesting. Here are the results.
>
> All run in interactive SQL, approximately 56M records. Index built
> over the numeric field (which took a freaking long time btw - like 10
> minutes).
>
> Using CvtDateToNumISO function with no index - 45 seconds
> Using function code with no index - 38 seconds
> Using CvtDateToNumISO function with index - 11 seconds
> Using function code with index - 2 seconds


This e-mail transmission contains information that is intended to be confidential and privileged.  If you receive this e-mail and you are not a named addressee you are hereby notified that you are not authorized to read, print, retain, copy or disseminate this communication without the consent of the sender and that doing so is prohibited and may be unlawful.  Please reply to the message immediately by informing the sender that the message was misdirected.  After replying, please delete and otherwise erase it and any attachments from your computer system.  Your assistance in correcting this error is appreciated.