Hi,
Sorry for the late reply, I have only just returned from holidays.
Please see below for answers to your questions:
But what is the limit of one instance then?
In relax there are no limits. If you have the memory available, then
relax will use it. If you go too far, the operating system OOM killer
will terminate relax (or on Windows and Mac OS, the system will start
swapping and become almost unusable).
Say I want to analyse the same dataset 100 times, each of the dataset have
experienced
an external parameter change. A normal analysis create 4/5 different pipes
for each
fitting model.
Would relax be able to handle 500 pipes?
No problems. If your computer will handle it is a different question.
Here I worry about the final_state file, saving, and memory management.
Is the memory released meanwhile?
This depends. If you call the pipe.delete user function between each,
then yes. Otherwise no.
Or should I flush/delete the pipes/memory for each 1/100 run?
If you run out of memory, then you will need to call pipe.delete. I
hope this helps.
Regards,
Edward
Re: Running with more that 1/2 CPU's