Cache de atributos oportunista: melhorando a eficiência do ABAC com o uso de uma política de distribuição de identidades em redes multinível para névoas computacionais

Abstract

Attribute-based Access Control (ABAC) is one of the most popular access control methods. Despite its popularity, a few works address attribute management in the Internet of Things (IoT). Most of the attributes needed for an IoT policy evaluation come from an external source. Therefore, managing attributes across the network requires communication between the policy decision point and the policy information point for each attribute, impacting ABAC performance. Attribute caches can mitigate this problem; however, due to the dynamic nature of the attributes, the cost of keeping caches up to date increases for each new replica. This work presents a method that predicts attribute requests and anticipates the attribute placement closer to the requester, balancing the cost of creating a new replica and giving ABAC performance benefits. The method handles the current attribute request and predicts where the subsequent request will occur. Based on simulations with a real dataset, the proposed method reduces above 80% the number of requests in the cloud using attributes’ caches and delivers up to 55% of the attributes in the first hop.