> I have not seen any statements of power efficiency on a transaction/watt > metric.
You haven't been to any analyst briefings in the last five years, then. Transaction processing capacity per unit power consumption has been a theme for a long time. IBM has several times put up graphics of various sorts discussing the problem. Simple extrapolation takes gate temperatures to rocket nozzle levels in the very near future. Advanced chip designs (perhaps even ECLIPZ) include concepts like powering off bits of the processor chip that aren't being used. Your company may never use Move Inverse or even floating point - if it takes three to five cycles to power these functions up when they're needed, why not turn them off after 100 cycles of non-use? However - sadly for this group - when IBM discusses transaction power efficiency it's MUCH more likely to mention iSeries (and pSeries) than zSeries. -- Phil Payne http://www.isham-research.co.uk +44 7833 654 800 ---------------------------------------------------------------------- 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
