I don't think that it is fair (nor easy) to compare distributed systems
to mainframes in an apples to apples style comparison. One was designed
to scale horizontally, the other vertically. Modern Unix and even
Intel/AMD based systems with fiber attached DASD can be rather
competitive in I/O comparisons and are clearly more suited for CPU
intensive operations. It's more about picking the right platform for the
job. It is damaging when we take a side either way and try to make it
sound as if one platform is superior in all respects to the other,
because this just is not true.

-Sam


-----Original Message-----
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of
Jon Brock
Sent: Thursday, June 30, 2005 1:36 PM
To: [email protected]
Subject: Re: Business Week Article


        My take on it is that, for 390/zSeries architecture, raw
computational performance takes a back seat to RAS; just about any time
during architecture design that there was a conflict between issues of
performance and, say, reliability, reliability would win.  To my mind,
that is the correct call.

Jon



<snip>
That's what I've heard as well. I've always wondered why the zSeries is
so computationally inferior to almost any other current CPU
architecture.
</snip>

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or
visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to