Forum Post: RE: Difference between logical reads and record reads

  • Thread starter Thread starter ChUIMonster
  • Start date Start date
Status
Not open for further replies.
C

ChUIMonster

Guest
Unless you use the new-fangled TABLE-SCAN option every record read involves the logical reading of both a data block and (at least one) index block. So the expected ratio of logical reads to record reads is a bit more than 2:1. (CAN-FIND() might, under some circumstances, result in index reads with no record reads.) 3:5:1 is a bit high -- but not scary high. It may indicate fragmentation, DBANALYS can give you some insight there. It might also mean that your index isn't as good as you think it is. Or that you aren't using the index that you think you are using. ProTop is very helpful here -- it will show you, at runtime, which tables and indexes are in use dynamically. The results sometimes surprise people. Also that "FIRST address OF customer" probably isn't doing what you think it is doing.

Continue reading...
 
Status
Not open for further replies.
Back
Top