Improving multiprocessor performance of a large telecommunication system by replacing interpretation with compilation

We consider different techniques for increasing the multiprocessor performance of an interpreted processing language in a large real-time telecommunication system, called Billing Gateway. We have implemented a prototype in which we first translate the language into C++ code, and then compile it using a C++ compiler. In our prototype we achieved a more than fourfold increase in throughput, compared to the original system, when running on an SMP with eight CPUs. The prototype also shows better scalability than the original system, due to less use of dynamic memory