A google search for "ns-2 free packet" revealed the link below, which I thought was interesting and possibly relevant to you. It has some suggestions for freeing packets. *How to scale your simulation to more nodes (500 nodes) and speed up it?* http://www.cs.binghamton.edu/~kliu/research/ns2code/index.html#speedup
On Tue, Mar 24, 2009 at 8:00 PM, <fh...@c3sl.ufpr.br> wrote: > > I am simulating a protocol which I developed, and I wanted the simulation > to end only when all my nodes are dead. Well, that is not really that fast > to happen, and before I reach that point my memory ends and I get a > bac_alloc error message. > > And here is the thing: I do alloc lots of packets with allocpkt, and I > also always manage to free them with Packet::free(Packet *), but even when > I free them, the memory usage don't even change ! And I am not > overreacting when I say this, I made some experiments, and the average > memory used when freeing all packet was exactly the same of when I did the > same simulation withou freeing them. > > > What do I need to do when I intend to have huge simulations, with like > 15000 packets flying in the air? Using Packet::free didn't solve my > problem. I even tried editing this function and inserting "free(pkt);" in > its end, but it was in vain. > > > Can someone please help me? > > > Thanks, > > -- > Fernando Henrique Gielow - UFPR - NR2 > Computer Science undergraduate student. > >
