Development of a parallel library for calculation of measures in complex networks

• Grant number: 16/08092-0
• Support type: Scholarships in Brazil - Scientific Initiation
• Effective date (Start): June 01, 2016
• Effective date (End): February 01, 2017
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Francisco Aparecido Rodrigues
• Grantee: Sady Sell Neto
• Home Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

The network theory has been developed since the end of the last century. Although several advances have been obtained, there are numerous challenges to be overcome. One of the most important is related to the limitation in the processing of networks made of million of vertices. Since the libraries currently available consider sequential processing, the calculation of some measures is very limited. In this project, we aim at developing several parallel functions for calculation of network measures. Basically, we will implement measures related to distance, centrality, spectral properties and community detection. These new function will enable the characterization of networks make up by dozen of millions of vertices, which can be done with supercomputer currently.