Performance Results
For performance and scalability testing I used 2 services: Intel Parallel Universe (PU) and Intel Multicore Testing Lab (MTL). MTL features machines with incredible 64 hardware threads which is good for scalability testing, while PU draws nice and comprehensible performance graphs and provides Intel Parallel Amplifier reports for quick dive into performance problems.
Here is a performance report from PU for (13,13) input:
And here are results of performance tests for 3 different machines: