|
One thing I konw - not assume - because I was at Rochester for a time last year in the lab, is that the internals of the Regatta processor _are_ different. I don't know everything - I was contracted through a vendor and therefore not privy to all details. I do know that the various processors, from Northstar on, have different levels of cache, different cache sizes, different ways that cache is managed - something called 4-way associativity may not be in the newer chip, e.g. One of the more recent chips had the capability to run in multitask mode. Idle time was used. Regatta does not, IIRC. Could this contribute to the discrepancy Steve sees? I seem to remember discussions down the hall, that there would NOT be a direct doubleing of performance. In another vein, capabilities may be known and may get implemented in a phased rollout. This happens in businesses all the time. Is this invidious? No. It lets you keep going in business. There is an argument that, if everything were done at one time, you'd not be in business in short order. At 07:04 AM 6/1/02 -0400, you wrote: >-----Original Message----- >From: midrange-l-admin@midrange.com >[mailto:midrange-l-admin@midrange.com]On Behalf Of Leif Svalgaard >Sent: Saturday, June 01, 2002 6:45 AM > > >The issue was: do you NEED special hardware to run fractional > >partitions. What does hardware provide for this that can't be done > >in software? This question has not been answered. > >It is a good question. > >I am still looking for the reason the cpw per processor of the new regatta >890 is only 40% more than that of the former high end 840. Even though the >clock speed is 100% greater. > >My calculations below. > >Steve Richter > > >( from the iSeries handbook, v5r1 ) >model 840, #2354. Cpw 15150 to 20200. Processors 18 to 24. Processor >speed: 600 Mhz. > >15150 / 18 = 841 CPW per processor. >20200 / 24 = 841 CPW per processor. > >Model 890. 37,400 CPW. 32 Processors. Speed: 1.2 Ghz ( as per press >reports ) > >37,400 / 32 = 1168.75 CPW per processor. > >1168 CPW = 140% of 841 CPW. > > > > >_______________________________________________ >This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list >To post a message email: MIDRANGE-L@midrange.com >To subscribe, unsubscribe, or change list options, >visit: http://lists.midrange.com/cgi-bin/listinfo/midrange-l >or email: MIDRANGE-L-request@midrange.com >Before posting, please take a moment to review the archives >at http://archive.midrange.com/midrange-l.
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.