get_complex_member

Hello,

I have written code that periodically takes specific number of samples (complex data) each time. I am traversing the received samples to collect dynamic data (read only) building a json object that will be sent to a db.
I am facing a series issue with network blocking/delay.

My assumption is that several concurrent threads (take thread and other conversion threads) are accessing the rti infrastructure resulting some sort of network blocking.

Further info:
I've got several threads:

1) Thread responsible for the take operation.

I have written code to traverse dynamic data to collect data (read only) of samples. The code yields the correct results, however, performance can be very slow. In particular, traversing samples with arrays of structures. Running valgrind, 70% of call time is dedicated to get_complex_member().  I have also tried bind_complex_member() which is 40% slower, and is counter to what I expected (eliminating the overhead of copying should be faster.)