25 Nov
2007
25 Nov
'07
11:56 a.m.
ok I do not remember but in objectmemory there is method that retrun
the size of an object (someone told me that it was a bit borken)
I do not remember its name
stef
On 23 nov. 07, at 15:30, Adrian Kuhn wrote:
That is an empirical value, I created 1 Mio anonymous
classes and
profiled memory footprint.
Exact value was 82.4568027 (which does not make sense as its not
modulo 4).
AA
On 23 Nov 2007, at 8:58 , Stéphane Ducasse wrote:
why F would be only 80 bytes?
you have superclass, subclass, format, methodDict + methods inside
(it does not have to have a metaclass) but may be you need it.
On 22 nov. 07, at 23:13, Adrian Kuhn wrote:
_______________________________________________
Moose-dev mailing list
Moose-dev(a)iam.unibe.ch
https://www.iam.unibe.ch/mailman/listinfo/moose-dev
On 22 Nov 2007, at 22:39 , Tudor Girba wrote:
_______________________________________________
Moose-dev mailing list
Moose-dev(a)iam.unibe.ch
https://www.iam.unibe.ch/mailman/listinfo/moose-dev I believe that here there are two points under
discussion, each
with
2 subpoints:
A1) the reasons for wide classes
A2) the solutions for wide classes
B1) the reasons for collective behavior
To save memory.
Each moose element has a back pointer to its model. This is 4 byte
per element, ie 4 MB per 1 Mio elements. Which are realistic maybe
even low numbers when analysing histories or traces. With collective
behavior the memory footprint is (4 + F) * M * T, where F is the
size
of an anonymous class, M is the number of models and T is the number
of element types used per model. Given that F is about 80 byte
(including object headers and metaclasses etc pp) and T maybe about
25, that is 2 KB = 0.002 MB per Model.
B2) the solutions for collective behavior
I propose to talk about each of these two points individually,
because they are not the same to me.
_______________________________________________
Moose-dev mailing list
Moose-dev(a)iam.unibe.ch
https://www.iam.unibe.ch/mailman/listinfo/moose-dev