The built-in multithreading capability of FLAMES^® allows scenarios to be executed using two or more of the processors (or cores) of a computer simultaneously. Using this capability will allow nearly all FLAMES scenarios to execute more quickly. If you want to execute larger scenarios in the same amount of time, or if you want to execute your existing scenarios in less time, multithreading may be just what you need.

When multithreading is enabled, FLAMES scenario execution is automatically divided into multiple threads (processes), and each thread is assigned a different portion of the scenario to execute. FLAMES and the operating system automatically distribute these threads to the various processors (or cores) of the computer and handle all shared memory management, thread synchronization, and load balancing.

FLAMES multithreading executes sensor detection, platform motion, and munition motion on multiple threads, because most scenario execution time for the majority of scenarios is spent in these operations. Other parts of scenario execution are executed on a single thread. If your computer has multiple processors or multiple cores, your scenario will usually execute faster, sometimes much faster. How much faster will depend on your computer and your particular scenario.

One of the most amazing and attractive features of multithreading is that it is almost always fully automatic. Changes to the application or the scenario are rarely required. The only requirement is that the models that are to be executed on parallel threads must be implemented according to FLAMES development guidelines.