OClient: On-board client clustering to bridge the interaction gap across multi-fabric adaptive ecosystem
The current blockchain paradigm is being hailed as an unquenchable goal to develop Hyperledger Fabric (HF)-based applications, resulting in a set of substantial enterprises shifting their networking operations, as intriguing consortia to plug the permissioned-nature of blockchain-based computing. While this practice endeavors to fundamentally revolutionize industries, one of the unavoidable problems that emerge is a lack of communication among autonomous HF-based networks, which can hinder the ongoing research landscape and consequently, have detrimental effects on its adoption. Towards this, blockchain interoperability (BI) has been heralded as the underlying concept by which communication can be derived through locating an inter-connection bridge. Considering BI’s significant traction, this paper introduces a new methodological approach to deal with the disparate nature of HF-based networks, termed OClient, for embracing two different interoperabilities (inter-network and inter-chain). This approach strategically tunes the inner working of clients by providing the global information of HF network topology that extends their conventional role as a matching service provider for facilitating opportunistic communication, in contrast to the existing solutions where typically third-parties have been exploited as the intermediaries. The paper streamlines the discussion by putting forth a pragmatic multi-HF (MHF) network scenario to formulate the disconnectivity problem, followed by constructing interaction-path-based graphs and conducting rigorous simulations. More importantly, this paper manifests experimental examples and corresponding results, providing insights across various dimensions of client-level interoperability so as to understand the impact on performance for combinations of inter-network and inter-chain scenarios. Succinctly, the results demonstrate that OClient is close to the performance of autonomous HF networks, which makes it a useful approach. Besides, this paper strives to understand how the adopted failure strategy to maintain ledger consistency contributes to performance and finally, concludes by anticipating significant future research directions towards HF interoperability.