>From what you've described it appears that this transaction can be divided up >into a certain amount of CPU time to process the transaction and then an >amount used to drive/manage the I/O. If that's the case, and the CPU consumed >for the transaction is relatively constant, you can approximate the cost of a >VSAM I/O by solving a pair of equations.
Let x = transaction CPU time and y = VSAM CPU time. Then by using two separate measurements (ie: 900 I/Os and 2000 I/Os), use the CPU service units as follows: x + 900y = CPU service (1) x + 2000y = CPU service (2) Subtract one from the other and solve for y. This will remove the constant for the transaction time and what is left over will be an approximation of what each VSAM I/O is using. From there, you can convert it to whatever units you like. I realize that this assumes that the transaction time also does NOT vary with the volume of data being read. But if that is true, then this should be a reasonable approach. Adam ---------------------------------------------------------------------- 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
