Martin, Based on the information provided, this is a bad idea.
1) It introduces SPOF (single points of failure). 2) A new datacenter should have its own customer number and the development MSUs should be reset to zero and calculated at the most expensive tiers (0 - 345), instead of at the cheaper clip level. 3) channel extension and teleco charges - don't they make it a bad idea? What exactly are they trying to accomplish? At my last company, I did the opposite of what you are trying to do...I merged a development datacenter into the primary. Saved a fortune in software $$$ and improved development and testing response times and batch throughput substantially. Not to mention, reduced network traffic. As to the distance question, my best guess is "it depends". What distance, who, what, where, when and how? Bob -----Original Message----- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Martin Kline Sent: Thursday, February 11, 2010 11:04 AM To: IBM-MAIN@bama.ua.edu Subject: GDPS Distance Impact on Sysplex Response Time We have a sysplex with four processors, two of which run production work with development work on the others. In their quest for trivia, my management wants to know the impact to production if we move the development processors to a second site, leaving the coupling facilities, DASD, VTS and primary network all on one site. I suspect this is a function of how much production data is being touched by the development systems as well as just the handshaking, messaging, etc. across the entire sysplex. I also suspect it affects some online transactions and batch work much more than other work. In addition, they want to know how much greater the impact is if the other site is twice as far away. And, just to make this interesting, they want an answer tomorrow. Actually they would like it yesterday, but tomorrow will have to do. I've been reading several manuals, and I think the answer they should get is 42. Does anyone have any suggestions on where to concentrate my efforts (only those that won't get me fired, please)? ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
