stein@dhw68k.cts.com (Rick 'Transputer' Stein) (12/13/90)
Suppose I could afford to use a cesium standard time reference for
the clock/timing source for every node in a MIMD-style massively
parallel computation system, would that guarantee that all instructions
were carried out in temporal synchrony provided that they we all calibrated
to the Naval observatories' clock?
Is the drift in the clock the sole cause of instruction slew, or is it
due to other silicon functions in the chip?
--
Richard M. Stein (aka, Rick 'Transputer' Stein)
Sole proprietor of Rick's Software Toxic Waste Dump and Kitty Litter Co.
"You build 'em, we bury 'em." uucp: ...{spsd, zardoz, felix}!dhw68k!stein