hadoop-tasks
Hadoop tasks repository for Parallel and Distributed Computing course at MIPT 2015
Contains code of the following tasks:
- Word Count
- Inverted Index
- Matrix Multiplication
Speed-up achieved for Matrix Multiplication
On 4-node Hadoop cluster Matrix Multiplication works for 1.5 min on 500x1000 and 1000x2000 matrices and sequential version of this program, written in Python, works for about 5 min.