Parallelizing Stream with Future

TL;DR

This paper introduces a method to parallelize stream computations by replacing lazy evaluation with futures, enabling concurrent execution of algorithms expressed as streams, and evaluates its performance on example algorithms.

Contribution

It proposes a novel approach to parallelize stream processing using futures instead of laziness, enhancing performance potential.

Findings

Parallelization improves execution speed of stream algorithms.

Performance varies depending on algorithm and implementation.

Discussion on methods to further optimize performance.

Abstract

Stream is re-interpreted in terms of a Lazy monad. Future is substituted for Lazy in the obtained construct, resulting in possible parallelization of any algorithm expressible as a Stream computation. The principle is tested against two example algorithms. Performance is evaluated, and a way to improve it briefly discussed.