On Mon, 8 Feb 2010 09:09:20 +0100, R.S. wrote: >Timothy Sipples pisze: >[...] >> Also, I'm quite sure that there's presently no cycle steering or other >> clock speed tricks to adjust power consumption dynamically, at least on the >> z cores. Considering the role mainframes play (running at high utilization >> rates), this design philosophy, to optimize for continuous operations, >> makes a lot of sense. > >Wrong. Cycle steering is present when REALLY needed. That means when >temperature grows up. It was discussed on the forum. Of course it covers >all the CPs. > So, no power reduction when a processor is in a wait state?
And capped general engines use no less power than specialty engines? What if _all_ CPs are capped? -- gil ---------------------------------------------------------------------- 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
