On Wednesday 19 December 2007 22:18, Jeff Woods wrote:
>
> After I wrote the above, a test indicated that if I raise RAM
> allocations to 384 MB around the clock, the estimated time to complete
> Stage 2 drops from 5.5 days to 1.5 days, so I've answered my own
> question for my own purposes... but given this, aren't the docs outdated
> when it comes to RAM requirements for P-1 factoring and/or the defaults
> for RAM allocation in the software such that most people won't be making
> effective use of their P-1 factoring time unless they both have the RAM
> to spare (as I do), and are educated enough to change the defaults?

I think there may be an issue with multi(core) processor systems here. The 
relative timings were AFAIK derived on uniprocessor systems. If you are 
running several instances of P-1 stage 2 in parallel you may well be cache 
thrashing to an unacceptable degree. However this situation will probably not 
persist as after the system has been running for a while the multiple 
instances will "desynchronise" so that it will be unusual for more than one 
instance to be running P-1 stage 2.

Setting memory to 8MB (which used to be the default) disables P-1 stage 2, but 
allows stage 1 (with a larger limit to compensate); this may be the short 
term answer to this "problem".

Better (in the long term) if an instance wants to start P-1 stage 2 but finds 
that another instance is already running stage 2, the subsequent instance 
could start LL testing immediately, then go back and run P-1 stage 2 when the 
other instance has finished.

Regards
Brian Beesley
_______________________________________________
Prime mailing list
[email protected]
http://hogranch.com/mailman/listinfo/prime

Reply via email to