zAAP was pitched as (and used for) offloading Java from the primary
engines. Neat idea!
It was several weeks, perhaps a few months, before I realized that the
zAAP was *not* a Java processor. Since it's all about microcode, I
figgered IBM had built a hardware JVM, a byte-code interpreter. Wow!
Sweet!! But no, just a ham-strung GP engine.
I was crestfallen.
We'd had IFL for four years already. (And I don't remember how long for
CF.)
Yep, fast development cycle for zAAP, but there was precedent for
capability limits. It's a slick trick and unique to Z class processors.
So ... it's really all about marketing and economics, little to do with
the actual tech. And it led to opportunities for third parties to fudge
things. (Which in turn led to the law suits.)
I understand the economics, but understanding and agreeing are not even
on the same axis.
-- R; <><
On 9/1/24 10:18 PM, Timothy Sipples wrote:
As previously reported back in 2004, the first customer production use of zAAP
(the System z Application Assist Processor) went live on September 1, 2004.
Which was impressively speedy because it occurred more than 3 weeks before the
earliest release of z/OS to support zAAP (z/OS Version 1.6) became Generally
Available — and barely 2 months after zAAP (the hardware feature) was
introduced. IBM discontinued zAAPs several years ago because their functions
were fully incorporated into zIIPs.
Happy 20th Birthday, zAAP!
—————
Timothy Sipples
Senior Architect
Digital Assets, Industry Solutions, and Cybersecurity
IBM Z/LinuxONE, Asia-Pacific
sipp...@sg.ibm.com
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
