On Tue, 14 Feb 2006 16:28:38 -0500, George Bly <[EMAIL PROTECTED]> wrote:
>There was talk of mainframe connectors? There were tools that could measure >mainframe stress(os, cics, disk, cpu) but no specifics. The mainframe is a >one engine 192mip we handle 1,767,000 transactions during an 8 hour day. >Plus batch, plus DB2, and plus TSO. We run 100% and we want to see where >bottlenecks occur while we ramp up web transactions. We use RMF to measure >things and had it all analyzed it shows we need CPU. My boss wants a second >opinion as to why and how many does it take. You might want to look at IBM's "IBM Tivoli Composite Application Manager for Response Time Tracking". http://www-306.ibm.com/software/tivoli/products/composite-application-mgr- rtt/ Haven't used it myself, but I saw a presentation at the zSeries Expo. Will show end to end response time with breakdowns for web server time, subsystem time (CICS & DB2) and network time. -Rob ---------------------------------------------------------------------- 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