Also curious about the "It also gives our end users the idea that z/OS is incapable of easy to use data access" remark, John.
If you're a keen or semi-keen observer of the IT world, relational databases are extremely popular and continuing to be popular, but non-relational databases (and data stores) are enjoying a robust renaissance. One size does not fit all. I think it's always a good idea to take a look at the full range of VSAM-related options: SYSB-II, VSAM Record-Level Sharing (RLS), and Transactional VSAM (TVS). And, to anticipate a question, no, you do NOT need multiple machines for either VSAM RLS or TVS. You don't even need more than one z/OS LPAR -- a monoplex is sufficient. You do need to define and to start a CFCC LPAR (or z/VM equivalent if applicable) if you don't have one already -- otherwise known as a configuration task, and all approaches require configuration tasks. That CFCC LPAR can either use (part of) a general purpose engine or a CF engine, and it needs a bit of memory allocated. The fact CFCC-related processing can run on a CF engine is a good, very zIIP-like option to have available because all these approaches incur some overhead. Whether it makes business sense to get a CF engine or not depends on how much CFCC-related processing you'll have. Often yes it does, but not always. The processing may grow with time as you use RLS or TVS more (and/or use your CF for other things) -- yes, new functions often get used and enjoyed -- so that decision can change over time, too. -------------------------------------------------------------------------------------------------------- Timothy Sipples GMU VCT Architect Executive (Based in Singapore) E-Mail: sipp...@sg.ibm.com ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
