Skip to content

Latest commit

 

History

History
10 lines (8 loc) · 1.04 KB

README.md

File metadata and controls

10 lines (8 loc) · 1.04 KB

Parallel Knn

The repo contains a C++ implementation of the KNN algorithm. There are four different implementations:

  • parallel.cpp: is a parallel implementation using only pthread. Input format: the value of K, the number of threads to use, the input file name, the output file name.
  • ff_parallel: is a parallel implementation consisting of a Master/Worker using the FastFlow library. Input format: the value of K, the number of threads to use, the input file name, the output file name, y if you want to use the custom thread pinning (is an adhoc implementation for the machine used during the experiments).
  • ff_parallel_for: is a parallel implementation consisting of a parallel for reduce using the FastFlow library.
  • sequential: is the sequential version of the KNN. Input format: the value of K, the input file name, the output file name.

In the experiments directory there are all the execution times computed during the experiments.
In the plot directory there are plots of the speedup, efficiency and scalability.