Multiprocessing - Geometric decomposition
Improving the Performance in the Statistical Redistribution of Message Symbols using Architectural patterns for Parallel Programming
1. Paper
2. Code
3. Results
The architectural design pattern for parallel programming called geometric decomposition helped us to reach the results we expected, reducing times by 52.28% for file 1 using 3 cores, and reduce by 28.20% for file 2 using 7 cores, these experiments help us satisfy most of quality attributes and avoid trade-offs that at first glance may seem impossible and at the same time satisfy most of the functional requirements It is important to highlight that the choice of the K factor is based on the one that offers the best compression, but when we choose a K factor very far from its consecutive one using large files perhaps the parallelism may degrade the performance. Sometimes it is a good decision to add some noise to the original message since this way you can get closer K factors, and this would be another topic of research.
Contributing and Feedback
Any ideas or feedback about this repository?. Help me to improve it.
Authors
- Created by Ramses Alexander Coraspe Valdez
- Created on 2019
License
This project is licensed under the terms of the MIT license.