Changes

Even though the runtime is closed after implementing the optimization, the kernels has a heavy load than before. It also finishes the accumulation of all randomly generated PI. I consider this optimization performs well.

Here is my final source code

|}