Wrote this scaffolding to practice MapReduce problems (simulated via processes).
python mapreduce.py
If you want to try it on different problems, you can modify:
dataset()-> input data.split()-> hostnames, paths, offsets, etc.map_func()reduce_func()
- The obvious non-distributed nature of this.
- Queues, poison pills, the output file, etc.
- Make it cleaner and more extensible. split, map, shuffle, reduce, combine, should be fill-in-the-blank boilerplate functions.
- Maybe add more datasets and problems to try.