GPU Accelerated Analysis and Simulation of Large Scale Networks.

Date of Submission

December 2015

Date of Award

Winter 12-12-2016

Institute Name (Publisher)

Indian Statistical Institute

Document Type

Master's Dissertation

Degree Name

Master of Technology

Subject Name

Computer Science


Advance Computing and Microelectronics Unit (ACMU-Kolkata)


Das, Nabanita (ACMU-Kolkata; ISI)

Abstract (Summary of the Work)

In the world of digital interaction, online social networks have become an inevitable part of our daily life. It becomes a strong and reliable medium to share individual views and other media to the rest of the world. In this report, we attempt to analyze, why some of the happenings in the world shared on the social network covers entire network and becomes popular or trending, while the others just die out over time or remains locally popular? Simulating or analyzing these in live environment poses big challenges due to their large size of underlying network and fast paced data generation. It is computationally intensive for CPU’s to manage such massive operations which is out of question as far as real-time results are desired. An efficient competitive diffusion scheme is proposed here to optimize for parallel execution in CUDA architecture. Extensive simulation study shows the viral and non-viral diffusion patterns and their associations with underlying graph structure.


ProQuest Collection ID:

Control Number


Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.


This document is currently not available here.