skerry, i am currently doing an analysis with 50000 equations which is
using roughly 50% of the available memory, according to "ps -aux". i
think this would include the physical memory (RAM), and not the swap,
but i'm not too sure. if its just the RAM, then 50% on my machine =
128MB. so for 270000 equations you're approaching a gig of RAM required.
(using large amounts of swap would take forever - i think its probably
not practical in your case, but someone else might think otherwise.)
your segmentation fault is a pretty clear indication that you don't have
enough memory. so you need to think about chucking some data out, or
using a non-reml procedure like method R. if you are going to chuck some
data out, my gut feeling would be to keep records from earlier years.
julius v.d.w. has i believe published some work on this sort of thing -
can't remember if it was in the same context though.
keep your hair on!
--
Andrew Swan
CSIRO Division of Animal Production
Pastoral Research Laboratory
Armidale 2350 AUSTRALIA
ph. +61 (0)2 67761377
fax +61 (0)2 67761333
email: andrew@chiswick.anprod.csiro.au