Thank you Greg, you have just verified my fears of all my research...sniff, sniff.. This is a simple select with a pass thru query from SAS, over the wire. It is 1.4 million rows, using a sorted_hash_cluster to sort the data for them so they won't do a group by and an order by in their query. The statement with both, without the sorted hash cluster takes an unknown amount of time- they just couldn't get it to complete. The sorted hash cluster completes in 4 hrs. Of course, now the boss wants to know- can I incorporate parallel? So much for that idea... We are pretty far into this design at this point, but any other ideas are always welcome! Thanks! Kellyn Pedersen Sr. Database Administrator I-Behavior Inc. http://www.linkedin.com/in/kellynpedersen www.dbakevlar.blogspot.com "Go away before I replace you with a very small and efficient shell script..." --- On Wed, 5/12/10, Greg Rahn <greg@xxxxxxxxxxxxxxxxxx> wrote: From: Greg Rahn <greg@xxxxxxxxxxxxxxxxxx> Subject: Re: Sorting Large Amounts of Data- Oracle to SAS, Part Deux To: kjped1313@xxxxxxxxx Cc: "oracle Freelists" <oracle-l@xxxxxxxxxxxxx> Date: Wednesday, May 12, 2010, 11:13 AM Parallel operations do not work on hash clusters and that is by design (so I'm told). Ten steps back... what is the technical requirement to have the data sorted? What is the vehicle that delivers the data to SAS, flat file, or simple select over the wire? On Wed, May 12, 2010 at 7:33 AM, Kellyn Pedersen <kjped1313@xxxxxxxxx> wrote: > OK, so I'm looking for some help with this one again- I've found a > compromise that has been quite successful for both groups so far: sorted hash > cluster. > Can I do it in parallel?? -- Regards, Greg Rahn http://structureddata.org -- //www.freelists.org/webpage/oracle-l