Calculation with chunks
This page shows the performance of the serial algorithm as the total_chunk_count is
increased from 1 to 120.
The performance of the simple dataset varies by only a small amount (2.5%) for a
total_chunk_count of up to 120.
For the random dataset the variations are greater at up to 13% and are usually, but not always,
slower for increasing total_chunk_count.