tinyield / sequences-benchmarks Goto Github PK

This is a workaround until Query.fromList works properly for this use-case

Why do we need 2 classes?

https://github.com/tinyield/sequences-benchmarks/blob/master/src/main/java/com/github/tiniyield/sequences/benchmarks/collapse/StreamCollapseOperation.java#L7

Analyze how many items must we have in a source stream to get advantage of parallel feature in a costumed Stream with the zip()operation.

To that end, we must provide a different zip() implementation able to partition both argument Streams with equal sizes. This one https://stackoverflow.com/a/46230233/1140754 should parallelize.

Despite some claims a bout the inefficiency of Stream parallel feature there are few evidences with benchmarks foundations. http://coopsoft.com/ar/Calamity2Article.html

The every and find pipelines are good use cases to explore how many items must we have in a Stream to get a speedup over the sequential processing.