ADSM-L

Re: [ADSM-L] Check signals on Power vs. x86...

2011-10-10 12:40:22
Subject: Re: [ADSM-L] Check signals on Power vs. x86...
From: "Allen S. Rout" <asr AT UFL DOT EDU>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Mon, 10 Oct 2011 12:24:43 -0400
Thanks very much for your various opinions; it's nice to have such a
good sounding board.


On 10/07/2011 11:20 AM, Shawn Drew wrote:

I think he was looking for Power vs x86 in price/performance.  I.E
If you spend 50K on Power systems and 50K on x86 systems.  which
could produce more I/O throughput.  (If not, that's what I'd like to
hear an update on)

Actually, I was aiming more for hardware/performance; this because
price:hardware can be very customer dependant.  I'm comparing, for
example, systems with X cpus of Y gigahertz, with Z core and Q
busses..

At a glance, it seems that the case is still strong for power, on
similar hardware numbers.

I'm trying to comb my bias out of the analysis, but I'm strongly in
Richard's camp:


On 10/07/2011 11:56 AM, Richard Sims wrote:

The major challenge these days is not computer cost but the cost of
supporting complexity, [...]


My various generations of AIX on power substrate under TSM have been
rock-solid for nearly 15 years now, and I've got 8+ TSM instances on
each system image.  So going 'wide' on less expensive, less performant
hardware seems a questionable win at best.



On 10/08/2011 03:46 AM, Daniel Sparrman wrote:

A Toyota Prius easily beats a Porsche in a price / performance
comparison. However, would you still buy the Prius?

I like the metaphor, but I think it erodes the case I want.  What's
the maintenance on a Porsche average in the first 5 years? :)



- Allen S. Rout

<Prev in Thread] Current Thread [Next in Thread>