Hi Kushagra, I agree that multi-threading in the core would be nice if it were able to improve execution time, but there are some hard problems to overcome to make that work in general. Here is some older work on that topic: [1]. Instead, ns-3 decided to focus on parallelization via MPI [2] (although the simulation problem being studied has to be amenable to this kind of parallelization) and on using multiple cores or systems to run experiments in parallel (e.g., the SEM tool [3]).
For Google Summer of Code, I suggest to review the project ideas listed by the prospective mentors and reach out to one of them to get started on developing an application.
- Tom
[1]. https://www.nsnam.org/workshops/wns3-2010/parallelization.pdf
--
You received this message because you are subscribed to the Google Groups "ns-developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ns-developer...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/ns-developers/59efed46-688b-446e-a859-35978adc7f19n%40googlegroups.com.