Hi there, we are having very slow respone times on a select statement. We have a file with a total of 5 million records in where the key is like nnnn-nnnnnn. The first part(nnnn) is a customer number(numeric) and the second part(nnnnnn) is a sequential no. We have created an i-descriptor on the customer number, FIELD(@ID,'-',1). On this i-descriptor we have put an index:
Index-Name...... F-type K-type Built Empties Dups In-DICT S/M F-no/VF-expr.... FAKT.FINR V Num Yes Yes Yes Yes S FIELD(@ID,'-',1) Selecting a customer with 570 000 records can take 30 seconds. (example of statement SELECT A_FILE WITH FAKT.FINR = 66) Selecting other indexed values in the same file which gives me 5 records out of the 5 million is very very fast. I've tried to rebuild.index with no improvment in performance. The file doesen't show any sign of level 2 overflow. I am single user on a powerful Sun server with SAN disks. We are on Solaris 8 and Unidata 6.1.10. Shouldn't it be faster than this? Perhaps this is normal and we have to take another approach to get the data we need? Thanks in advance! Bjorn Eklund Faktab Finans AB ------- u2-users mailing list [email protected] To unsubscribe please visit http://listserver.u2ug.org/
