I'm not a UD user, at least for quite a while, but isn't there a UD file type where the ID is always stored separately from the data, so that a select based soley on id-dependent criteria alone would be very fast because the actual data isn't read. If I'm right & if that's what you're using, I think you need to separate out not just the selection, but also the reading & building of the arrays and only start the timing test on the extractions. Or maybe that's what you meant.
> -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] On Behalf Of Bill Haskett > Sent: Friday, March 02, 2007 10:11 PM > To: [email protected] > Subject: RE: [U2] Universe vs. Unidata > > Bruce: > > The difference is just as startling regardless of which test > is done first. It's the array processing that kills the > timing, now whether the data is already cached (the select > only takes a few seconds either way). > > The timing doesn't start until the data is already selected. > > Bill > > > |-----Original Message----- > |From: [EMAIL PROTECTED] > |[mailto:[EMAIL PROTECTED] On Behalf Of Bruce Nichol > |Sent: Friday, March 02, 2007 6:58 PM > |To: [email protected] > |Subject: RE: [U2] Universe vs. Unidata > | > |Goo'day, Bill, > | > |If I can make observations and suggestions here? > |Observations: > |1. The time (5:30pm) indicates to me a quiescent > machine.... no other > |activity? > |2. The second run in each case would almost certainly involve items > |being in memory, although that would not be the case for the > first run. > | > |Suggestion: > |Do the same thing on a quiescent machine and reverse the process > |(ReadnextLoop first, followed by ArrayLoop) > | > |There may well be a difference, but I don't think it'd be so > |startling.... > | > | > |At 17:54 02/03/07 -0800, you wrote: > | > |>Chuck: > |> > |>Here is a UniData test run... > | > |Regards, > | > |Bruce Nichol > |Talon Computer Services > |ALBURY NSW 2640 > |Australia > | > |http://www.taloncs.com.au > | > |Tel: +61 (0)411149636 > |Fax: +61 (0)260232119 > | > |If it ain't broke, fix it till it is! > |------- > |u2-users mailing list > |[email protected] > |To unsubscribe please visit http://listserver.u2ug.org/ > | > ------- > u2-users mailing list > [email protected] > To unsubscribe please visit http://listserver.u2ug.org/ ------- u2-users mailing list [email protected] To unsubscribe please visit http://listserver.u2ug.org/
