On Fri, Aug 15, 2003 at 09:01:32AM -0400, Jon LaBadie wrote:
> On Fri, Aug 15, 2003 at 09:06:41AM +0100, jason.walton AT nomadsoft.co DOT uk 
> wrote:
> > Run Time (hrs:min)         5:42
> > Dump Time (hrs:min)       19:34      19:34       0:00
> 
> [...] if there were no parallelism in the phases that [run time] would
> be approximately the sum of the three phases.  But never
> shorter than any of the single phases like your "dump time"s
> each much longer that the total "run time"s.

If I'm not mistaken, "dump time" gives the sum of the durations
of all the dumps.  In other words, it's how long all the dumps
would have taken if they hadn't run in parallel.  (More or less
-- if they hadn't run in parallel, they'd probably have taken a
different amount of time due to a different resource-usage
pattern.)

So the lower the runtime:dumptime ratio, the more time
parallelism is saving you.

--

|  | /\
|-_|/  >   Eric Siegerman, Toronto, Ont.        erics AT telepres DOT com
|  |  /
When I came back around from the dark side, there in front of me would
be the landing area where the crew was, and the Earth, all in the view
of my window. I couldn't help but think that there in front of me was
all of humanity, except me.
        - Michael Collins, Apollo 11 Command Module Pilot