How a decentralized peer-to-peer based private contact discovery system performs depending on user base size and network performance

Detta är en Kandidat-uppsats från KTH/Skolan för elektroteknik och datavetenskap (EECS)

Författare: Martin Fagerlund; Adam Melander; [2021]

Nyckelord: ;

Sammanfattning: Internet privacy and security has become more discussed in public in recent years. Decentralization is a technique that removes trust in central authorities by distributing authority across a network of nodes. But it comes with challenges. This report investigated the advantages and limitations a decentralized contact discovery system for a messaging application may have. The areas of interest in this study concerned performance and scalability of the system. The simulation used a mathematical approach where each task's expected runtime was estimated from experiments or previous conducted studies. Multiple tests were performed to examine how different factors affected the overall performance. The results from the simulations showed that the performance and scalability of the system could be considered viable. However, since the simulation couldn't be compared to a working system or previous research and therefore lacked an estimation of its accuracy together with some limitations in the model, further work needs to be done in order to determine the simulation's predictive power. 

  HÄR KAN DU HÄMTA UPPSATSEN I FULLTEXT. (följ länken till nästa sida)