A Multi-Functional Content Distribution Framework Centered on Edge Authentication

Abstract

The paper proposes a framework solving problems of traditional
centralized data distribution architectures. Those architectures
suffer from high server load, high latency and tightly bound data
and authentication logic. The goal is not only faster computation,
but also smarter distribution and management. It takes advantage
of edge computing and separates the outermost layer of security
control from traditional servers. That can prevent origin servers
from being exposed to the public network. Small and medium
scenarios such as school clubs and public facilities can benefit from
this solution. It works like a small server, and can reduce server load
by processing simple requests like invalid requests, high-frequency
AI data collection and suspicious requests. The framework features
distributed token check, dynamic routing and multi-scenario data
distribution. The framework needs a platform key, database and
customized routing rules. It can also run without a database. The
framework can be enhanced by adding custom extensions. This
framework is applicable to dynamic password distribution (e.g.,
offline software authentication), limited data distribution (e.g.,
school club file sharing), IoT device authentication (e.g., smart
sensors in classrooms), or serving as an API gateway (e.g.,
transforming map APIs to a unified format).