Prelude/SX Enterprise and UniData
Greetings, My company is looking at acquiring one of two ERP systems: Prelude and NxTrends' SX Enterprise. Ardent's UniData was recommended as one possible back-end and we are in the process of acquiring benchmark information to get an idea of its performance compared to other RDBMSs. Has anyone utilized either ERP with this back-end and how has it performed? Any feedback is deeply appreciated. Thanks much, Matt Zuberko Database Analyst Travers Tool Co. Flushing, NY Sent via Deja.com http://www.deja.com/ Before you buy.
I am a informix online dynamic workgroup server user .I wanted to know the formula for cost of query ,when we run a set explain on; on a query <zuberkom@my-deja.com> wrote in message news:8mmhi5$32p$1@nnrp1.deja.com... > Greetings, > > My company is looking at acquiring one of two ERP > systems: Prelude and NxTrends' SX Enterprise. > Ardent's UniData was recommended as one possible > back-end and we are in the process of acquiring > benchmark information to get an idea of its > performance compared to other RDBMSs. Has anyone > utilized either ERP with this back-end and how > has it performed? Any feedback is deeply > appreciated. > > Thanks much, > Matt Zuberko > Database Analyst > Travers Tool Co. > Flushing, NY > > > Sent via Deja.com http://www.deja.com/ > Before you buy.
pankaj wrote: > I am a informix online dynamic workgroup server user .I wanted to know the > formula for cost of query ,when we run a set explain on; on a query To the extent that there is an answer, it is described in the manuals available from http://www.informix.com/answers. The formula varies slightly from release to release, and SET EXPLAIN only tells you the answer it chose and nothing about the answers it rejected. The whole business is extremely complex if you have distributions involved. And you need to run UPDATE STATISTICS correctly (use do_stats from Art Kagel's utils2_ak from the software archive at the IIUG web site, http://www.iiug.org) Fundamentally, you can assume that lower numbers are better/cheaper. The relative magnitudes of the numbers give some indication of which alternative query plans (or parts of a query) are most expensive. The whole game is somewhat fraught with finger-in-the-air guesses. Classically (in the absence of distributions), some filter conditions were assumed to select 1/10 of the rows; this could be horrendously off target, depending on the nature of the condition and the data against which it was actually run. -- Yours, Jonathan Leffler (Jonathan.Leffler@Informix.com) #include <disclaimer.h> Guardian of DBD::Informix v1.00.PC1 -- http://www.perl.com/CPAN "I don't suffer from insanity; I enjoy every minute of it!"