It definitely played its role in high lighting what most functionish coders already knew: that map/filter/reduce is an awesome model for data processing pipelines.
Huge caveat to this: systems like Hadoop accomplish the parallelization by adding keys, and those FP constructs don't broadly have this. It's better to think of it as parallel divide-and-conquer.
Comments
It definitely played its role in high lighting what most functionish coders already knew: that map/filter/reduce is an awesome model for data processing pipelines.
Huge caveat to this: systems like Hadoop accomplish the parallelization by adding keys, and those FP constructs don't broadly have this. It's better to think of it as parallel divide-and-conquer.