Distributed Ledger Technologies for Managing Heterogenous Computing and Sensing Systems at the Edge

Abstract

The increased popularity of Internet of Things (IoT) devices, ranging from simple sensors to powerful embedded computers, has created the need for solutions capable of processing and storing information near those assets. Edge Computing (EC) has become a staple architecture when designing solutions for IoT, as it optimizes the workload and capacity of systems dependent of the Cloud, by placing the required computing power near to where the information is being produced and consumed. An issue with these solutions, is that reaching consensus regarding the state of the network becomes more challenging as they scale in size. Distributed Ledger Technology (DLT) can be described as a network of distributed databases that incorporate cryptography and algorithms to reach consensus among the participants. DLT has gained traction over the past years, particularly due to the popularity of Blockchain, the most well-known type of DLT implementation. In addition to the capability of reaching consensus, another key concept that brings EC and DLT together, is the reliability and trust that the latter offers through transparent and traceable transactions. In this thesis, we present the design and development of a proof-of-concept system that uses DLT Smart Contracts (SC) as the core for efficiently selecting Edge Nodes for offloading services. We present the experiments conducted to demonstrate the efficacy of the system and our conclusions regarding the usage of Hyperledger Fabric for managing systems at the edge.