If one were looking at CPU, though, it might be a different matter... Long long ago in a decade far far away was The Coffee Table Book. :-) Or so I called the Data In Memory studies orange book. It showed the results of a number of performance studies conducted by Poughkeepsie (and possibly other labs). Some of the fine people who did the studies may very well be watching IBM-MAIN. :-)
One of the things I remember is that over a number of data points the VIO to Expanded Storage study showed a net CPU cost for using VIO to ES over regular temporary data sets. The folklore suggested that this was due to simulating a real (then 3380) device via VIO. Recall it maintains the illusion of PHYSICAL I/O to real disk volumes. Now, things may have changed. In particular it's not Expanded Storage but Central. But I doubt the device simulation aspect has changed much. So I would be cautious on CPU for VIO to Central against SSD. Anyone who knows anything SUBSTANTIAL like to chip in at this point? Martin Martin Packer Performance Consultant IBM United Kingdom Ltd +44-20-8832-5167 +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter ID: MartinPacker "They're figuring out that collaboration isn't a productivity hit, it makes them smarter." Sam Palmisano on BlogCentral, 26 November 2008 Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU ---------------------------------------------------------------------- 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