On Wednesday, 05/17/2006 at 12:42 AST, John Campbell/Tampa/[EMAIL PROTECTED]
wrote:
> #1: A mainframe is designed to provide maximum *reliable* single-thread
> performance (per CP) simply because a lot of the business workload
(merge
> phase of sortation, for instance, or balancing a b-tree after an insert)
> needs it and errors can't be tolerated.

And yet I've seen graphs showing how z is optimitized for
multiple-workload (thread) performance because of its low-cost context
switching.  That's why our memory and cache designs are the way they are,
and why we can run multiple applications *well* on a single CPU.

> #2:  A mainframe provides maximum I/O connectivity.  Granted, as we move
> towards fibre channels we're getting away from this differentiator.  I
wish
> I had a Shark ESS to play with...

But there's still the sheer volume of connectivity: The old z9-109 allowed
up to 84 4-port FICON Express2 cards for a total of 336 FC ports.  (Truth
in advertising: OSAs, ESCON, FICON, and Crypto cards all compete for those
84 I/O slots.)

Alan Altmark
z/VM Development
IBM Endicott

----------------------------------------------------------------------
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