With the ability to use multiple CPU cores, we need to make sure there is no degradation of scalability => the same scenario should reach a given score in a certain amount of time. The amount of time should be different for 1,2,4 threads, but for the same number of threads, the time should also remain the same over time. Also, the amount of time should not be worse than for a single-threaded solving.

Create a scalability test for multi-threaded solving on solutions with chained and non-chained variables (e.g. CloudBalancing and VehicleRouting).

What to measure and record:

• execution time
• memory consumption

Measured values must get into performance results repository.

Updating performance report with these new metrics will be tracked in a separate ticket.