16 May
2024
16 May
'24
5:07 a.m.
I thought the CDC CYBER and 6000 series mainframes were great systems which
performed admirably for what they were designed for. I liked COMPASS, SYMPL
and NOS 1 and 2. I didn't do much work in CYBIL, but it was basically an
enhanced version of PASCAL suitable for operating systems work. What is
there not to like? These mainframes and the CDC 7600 outperformed every
other machine until Seymour Cray released his own machines.
On Thu, May 16, 2024 at 11:08 AM Ken Seefried via cctalk <
cctalk(a)classiccmp.org> wrote:
I came to it all a bit later. I do recall the CDC
salesthing saying
something like "oh, you guys have some Unix around here? Have we got
something for you!". And the systems guys brought up NOS/VE on the last
CDC machines we ever bought.
On Wed, May 15, 2024 at 10:43 PM Chuck Guzis via cctalk <
cctalk(a)classiccmp.org> wrote:
On 5/15/24 18:47, Ken Seefried via cctalk
wrote:
Please...I'm trying very hard not to remember
them (or NOS...worse,
NOS/VE).
I left CDC at around the time that SCOPE 3.4 was being renamed NOS BE
and KRONOS was becoming NOS. I remember attending a design meeting for
the pager in what was to become NOS/VE. I asked the presenter if he'd
conferred with any of the virtual memory pager talent that CDC had
in-house. Blank stare. I informed him that the STAR-100 people had
lived in that particular hell since about 1969--and that demand paging
was not the way to run a shop. STAR had long-since switched to a
working set algorithm.
Even that wasn't enough. If one selected a large (65 KW) page size and
set up certain vector instructions so that addresses crossed page
boundaries, it was impossible to get the required pages into memory all
at once.
The system just sat there and thrashed....
--Chuck