[strongSwan] Locking contention or delay in the Charon process under high load

[Martin Willi]

Chinmaya,

> Using the pthread_setaffinity_np() API to put threads into different
> cores, I find the tunnel setup rate to be 400+ (maximum) without any
> packets loss at both ends. Without setting processor affinity, only
> once core gets used (100%) and setup rate was found to be 250 (max). I
> think, it helps to improve the locality of memory access, performing
> load balancing and achieve parallelism. 

> I am totally stuck, unable to proceed further. Kindly give me some
> clues to move forward. 

Sorry, I don't have the time nor your hardware available here to dive
deeper into the issues you are experiencing. I can't assist you beyond
what I did for these specific issues on your very specific hardware.

Upscaling to this level on this kind of hardware is certainly
non-trivial, but probably you'll have to analyze and optimize your
specific situation yourself to achieve the setup rate you are targeting
(if possible at all).

Kind Regards
Martin