Re: A personal view of SigAda

[jpwoodruff@irisinternet.net (John Woodruff)]

JeLlyFish.software@gmx.net (Vinzent Hoefler) wrote in message news:<36f61440.0212201345.74a04ba3@posting.google.com>...

> > The keynote from Robert Carey of Livermore Labs was amazing - they
> > are building a thing called the "National Ignition Facility" (NIF),
> > which is basically one of the world's biggest lasers.
> 
> Oh yes, the figures looked very impressive...
> 
> > Oh, and when the shot fires, the whole thing is supposed
> > to be synchronised to <20 _pico_seconds!  Now that's what I call
> > hard real-time!
> 
> I still wonder how they will achieve that, especially on a relatively
> large distributed system like this. I mean it basically requires more
> accuracy than the CPU clock alone has...

Here's a little clarification about the NIF's 20 picosecond timing
precision:

Of course, there is no computing taking place during the propagation
of the laser pulse (which has duration 25 nanoseconds).  The
Integrated Computer Control System that Bob described sets up the
experiment and then turns over control to a hardware timing system and
that hardware distributes triggers to the 1500 or so instruments that
refine and observe the pulse.

"The Integrated Timing System uses fiber-optic communications and
network technologies to distribute a time-encoded data stream
throughout the facility; then generates independently controlled,
real-time trigger pulses close to the device to be triggered."

A report "The NIF Integrated Timing System - Design and Performance"
by Dick Lerche et al was delivered at the ICALEPCS 2001 conference and
is available at http://www.slac.stanford.edu/econf/C011127/TUAP069.shtml.