Regarding #1: (9.2.0.4 Solaris 9) We have a recommendation from the sw vendor to compute with histograms using exec dbms_stats.gather_schema_stats(ownname => 'SCOTT', estimate_percent => dbms_stats.auto_sample_size, method_opt => 'for all columns size skewonly', cascade => true); as an initial method to gather the stats. I did a bit of looking and a bit of testing, and it looks good to me. Down side: it did take several hours to gather the initial set of stats. What are other folks doing with histograms?? Barb --- Mogens_Nørgaard <mln@xxxxxxxxxxxx> wrote: > Can't remember if you got any responses on this - if > you did, then I > apologize for any repetitions... > > There's no correct answer to this question - if > there was, it would be > part of the DBMS_STATS thing now, I bet :-) > > I would suggest the following (subject to debate, > please): > > 1. Collect stats (sample 1 percent) on all your > objects OR > 2. Import stats from your test system (or another > similar system) ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx put 'unsubscribe' in the subject line. -- Archives are at //www.freelists.org/archives/oracle-l/ FAQ is at //www.freelists.org/help/fom-serve/cache/1.html -----------------------------------------------------------------