Carlo de Wolf wrote:
First I want to see some measurements to know the extend of the problem and to have a base of comparison.
I agree. In looking at the code, all the action is in ServiceContainer and InvocationStatistics. I will do some timings on the baseline, and then remove the invokeStats object completely, so I can see a delta. To get enough concurrency to make it reasonable, I might borrow a large maching in the RHTS, so we could measure with 100's or even a 1,000 concurrent threads. So, I'll try to get some comprehensize measurements, so we can quantify the overhead.