Performance Analysis of a Blockchain Based Peer-to-Peer Energy Trading Framework

Abstract

Peer to Peer (P2P) energy trading were introduced to store, manage, share, track, and access the permanently stored energy data. In particular traditional energy trading systems provide a centralized approach and connectivity among the different participating trusted users. Recently Blockchain Technology shows a potential solution to address critical challenges such as gapped communications, security in an untrusted decentralized environment, inefficient transactions retrieval (low throughput, high latency, high CPU and memory utilization, etc.), and scattered energy records. In this paper a framework is proposed, to record the energy transactions, a JavaScript-based smart contract is deployed over the Hyperledger Sawtooth Proof of Elapsed Time (PoET). Additionally, the proposed framework is shown as a working prototype, and the performances are evaluated taking transaction per second, read and write latency, read write throughput, the number of successful and failed transactions, CPU utilization, network traffic input, and output. Also, it is shown that PoET's performance is better than PoW for multi-hosting environments. It has been observed that out of 1000 transactions sent for each round by clients through Hyperledger Caliper and REST API services, PoET could able to perform 946 transactions successfully, whereas PoW completed only 627 transactions. The results obtained in various rounds of evaluation affirm the superiority of the framework.