On 13 Apr 00, at 7:34, Martijn Kruithof wrote:
> This happened on my machine while switching from beta 3 to beta 4 is
> this correct?
>
> Starting P-1 factoring on M33237229 with B1=330000, B2=2640000
> Chance of finding a factor is an estimated 4.08%
> P-1 on M33237229 with B1=330000, B2=2640000
> M33237229 stage 2 is 12.08% complete. Time: 1822.421 sec. (789108274292
> clocks)
>
> And after installing beta 4: (Stage 1 was not redone!)
>
> Starting P-1 factoring on M33237229 with B1=305000, B2=1525000
> Chance of finding a factor is an estimated 3.58%
> P-1 on M33237229 with B1=305000, B2=1525000
> M33237229 stage 2 is 20.62% complete. Time: 1767.363 sec. (765268138179
> clocks)
Looks a bit odd. But if you're _reducing_ the limits it doesn't seem
to be neccessary to recompute Stage 1, you actually have a bit more
info than you need to proceed straight to Stage 2.
I wonder if RollingAverage slipped a bit, causing the compute time
estimate to increase & resulting in lower limits. George, surely this
is wrong. RollingAverage correction applies identically to the P-1
factoring time and the LL testing time, so the bounds should not vary
just because the effective system speed estimate does.
> Furthermore a feature would be very nice that it does not recalculate the
> bounds without the users consent. (even if it violates the memory bounds)
I don't know why the bounds were recalculated in this case. But
busting the memory bounds is serious.
>
> The way to run the memory constraints on a dual machine is by the way
> configuring day/nighttime differently for the two cpu's as the stage 2 is
> only very short, it does not matter if that's only 4 hours (it can resume
> the next day anyway). (i.e.) one has daytime from 6 pm to 6 am, the other
> from 6 am to 6 pm. (real daytime from 19 to 23 pm might yield : daytime
> from 19 to 6 and one daytime from 6 to 23)
Yes, that's one way of doing it. But, if you have them both set to
the same time window (& given that 2*NightMem is comfortably less
than the installed memory less whatever is claimed by the OS etc)
things work OK, P-1 stage 2 just waits until night time comes around.
What worries me about your approach is that, if you have a Stage 2
running in "night time" on processor 1, then the processor day/night
switches over and a second Stage 2 starts on processor 2, the Stage 2
job on processor 1 will continue to completion. If you have set
2*NightMem bigger than the available system memory, your memory will
suddenly be overcommitted; the disk activity LED will lock on and the
system will start to run very slowly indeed.
Regards
Brian Beesley
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
