Yes the times are in uses of 128 microseconds. 

If you are a member of CMG or are willing to register on their website you can 
find my 1998 paper "DOES YOUR INSTALLATION HAVE HALF-SECOND I/O RESPONSE TIME? 
(A Closer Examination of Disconnect Time)". I believe that paper first explored 
how to use the SMF 42-6 data to better understand response time for cache hits 
and cache misses.  

Hard to believe that was 15 years ago that I did that work. 


Al Sherkow

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to