Hi Ted: For what is the 60% buffer hit ratio? Is it for the data? Is it for the index? Or is it for both? If you have a 3 level file, and you are only getting a 60% hit ratio, then it you are doing an average of 1.6 I/Os per request. From a bird's eye view, you go from the application to CICS management modules, to VSAM to the SVC handler in the supervisor, to IOS to issuing the SSCH and then going back up the chain to wait until the record is read (which in this case may be index record and requires this operation to be re-issued until the data is read). Once the I/O interrupt occurs, you have to handle the interrupt, create/schedule the SRB, execute the SRB in the region, get the CICS dispatcher to re-dispatch you and finally you continue in your application. If you had gotten a look-aside hit, most of the instructions executed would not have occurred because you would have returned from VSAM to CICS.
So, a savings of 10K instructions (I/O cost minus look-aside cost) is feasible. Also, you don't have a very high cost of waiting for I/O as you are getting a 98% cache look-aside but you still went through the instructions in the operating system for the I/O. The data table is an excellent solution if you have real storage to back the file definition because you eliminate the index search and hash to the record desired. Regards, Gene ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
