virtual memory 4503

On Thu, 01 Jun 2006 17:07:33 -0400 in alt.folklore.computers, Bill

The comment was meant to be general about various papers in that era doing studies which read as advocacy for a theory rather than a critical examination of the empirical basis for the theory.

Kind of hard to believe results claimed in studies which one has not read, is not publicly available, and whose baseline can not be explained by those quoting it.

Reference to design by Cutler, and what appears to me IMHO to be what BAH calls small computer thinking: producing another small system when we customers wanted systems larger than PDP-11-70s, and DEC were refusing to sell multiprocessor 11s or DEC-20s.

Reference was to overhead in dropping and reclaiming 512B pages, whereas 4-8KB pages would have been reasonable sizes.

BTDT with TKB under RSX and RSTS-E emulation, also with RT11, DOS-11, and other DEC linkers. Always liked DEC's overlay management code: you come up with a feasible structure, and it'll take care of you; screw up your structure or sizes, and wild pointer chases were easy by comparison.

-- Thanks. Take care, Brian Inglis Calgary, Alberta, Canada

fake address use address above to reply