Update Stats slowing.....

IDS9.30HC5, HP-UX 11i 4cpu Morning all! Our overnight update stats is slowing up on us on one table - should it be taking over 3 hours to run UPDATE STATISTICS LOW on a table with 40,000,000 rows (85 columns, no special columns (text, byte etc), 23 indexes, 2 fragments and 29 extents) in overnight quiet time? OK I know I need to address the extenting issue but the thing is the table hasn't really changed greatly except for day-to-day growth of about 500 rows over the last year or so; it's just suddenly gone up from an hour or so. (We do MEDIUM DISTRIBUTIONS ONLY for the whole table beforehand and it takes 6 minutes!) We have MAX PDQ set to 10, PSORT_NPROCS=4 and DBUPSPACE=10000. PSORT_DBTEMP is not used. Any input gratefully received. Cheers Malc

Malcolm, I don't think the problem is anything to do with extents. We have many tables with over 100 extents and quite a lot of indexes and we never see times like that. In fact our largest site does UPDATE STATS MEDIUM on the whole database faster than that. But have you looked closely at the indexes. Are they detached? If so, how many extents do they have? Could it be skewed indexing? Also, are there any errors when you access the table and/or indexes? I'd think about running oncheck against the complete table. Just a few thoughts The Original Malcolm - all others are merely copies. -----Original Message----- From: informix-list-bounces@iiug.org [mailto:informix-list-bounces@iiug.org] On Behalf Of malc_p@btinternet.com Sent: 17 March 2006 10:41 To: informix-list@iiug.org Subject: Update Stats slowing..... IDS9.30HC5, HP-UX 11i 4cpu Morning all! Our overnight update stats is slowing up on us on one table - should it be taking over 3 hours to run UPDATE STATISTICS LOW on a table with 40,000,000 rows (85 columns, no special columns (text, byte etc), 23 indexes, 2 fragments and 29 extents) in overnight quiet time? OK I know I need to address the extenting issue but the thing is the table hasn't really changed greatly except for day-to-day growth of about 500 rows over the last year or so; it's just suddenly gone up from an hour or so. (We do MEDIUM DISTRIBUTIONS ONLY for the whole table beforehand and it takes 6 minutes!) We have MAX PDQ set to 10, PSORT_NPROCS=4 and DBUPSPACE=10000. PSORT_DBTEMP is not used. Any input gratefully received. Cheers Malc _______________________________________________ Informix-list mailing list Informix-list@iiug.org http://www.iiug.org/mailman/listinfo/informix-list