PROJECT TITLE :

Resource Analysis of Blockchain Consensus Algorithms in Hyperledger Fabric

ABSTRACT:

Because the consensus algorithm is used in the blockchain network to tolerate node faults while maintaining the consistency and integrity of data, it is an essential component of all blockchain services. Previous research on the consensus algorithm has had some limitations, including the following: 1) There was no analysis of the consumption of resources carried out, 2) the performance analysis was not exhaustive in terms of the parameters of the blockchain (such as the number of orderer nodes, the number of fault nodes, the batch size, and the payload size), and 3) there was no evaluation of the potential outcomes of practical fault scenarios. To put it another way, the issue of adequately addressing the provisioning of resources for consensus algorithms in cloud environments has not been resolved. The question of how to provide consensus algorithms becomes an important one as more blockchain services are implemented in the form of blockchain as a service (BaaS). This question needs to be answered. A kernel-level analysis for the resource consumption as well as comprehensive performance evaluations of three major consensus algorithms are presented in this study (i.e., Kafka, Raft, and PBFT). Our experiments have shown that the amount of resources used can vary by as much as seven times, which highlights how critical it is to ensure that BaaS is adequately provided with resources.