Performance Evaluation of Decentralized Social Media on Near Protocol Blockchain

Abstract

Social media's decentralization gives users control of their data, contributing to privacy, ownership, and dissemination. The challenge is to develop a decentralized social media platform with the same features and performances as its centralized counterpart. This study aims to demonstrate and evaluate the performance of decentralized social media in terms of throughput and scalability. The prototype application is developed on NEAR (Near Protocol Blockchain), a non-EVM (Ethereum Virtual Machine) chain with sharding. The results show that the use of NEAR to store post content is not a scalable solution, and the workaround of using IndexedDB as a local database to store information reduces response time and increases throughput and scalability. Collaboration of on-chain and off-chain storage strategies performs better for decentralized social media applications.