Scaling Hyperledger Fabric via Blockchain splitting

Over the last decade blockchain platforms have attracted increasing interests; the maturation of the technology allowed it to develop from a peer-to-peer payment system originally inspired by the cypherpunk movement to a technology which today also finds applications in the industry. This evolution required an adaptation to fit with the needs of enterprise use, thus causing the transition from the permissionless paradigm of the original open blockchains to the permissioned one, more suited to the use case. However the technology is not yet ready for industrial massive scale deployments, since it presents some scalability problems due to the underlying Byzantine Fault Tolerant protocols used to reach consensus on the order of transactions, which hinder its adoption. Several different approaches have been proposed by the research community to mitigate this problem, some directly aiming at improving the scalability of consensus protocols, others trying to tackle the problem from an alternative perspective, i.e. by setting up off-chain channels to lower the workload of the actual blockchain. In architectural terms, our proposal positions itself in the middle of these two extremes, beside sharding approaches. Indeed we present a solution which aims at scaling permissioned blockchains through adaptive parallelization, more specifically by dividing individual blockchains into smaller child blockchains when they become too large, thus ensuring peak performances at all times. Our solution aims at achieving this objective while still ensuring maximum flexibility and autonomy to the individual blockchains, thus effectively contributing to the state of the art. In order to efficiently achieve this goal it is of crucial importance to provide means for interoperability between individual blockchains, therefore we include in our solution two cross chain communication protocols, namely Transfer of Knowledge and Transfer of Asset, which allow for trustless flows of information and assets across blockchains. Furthermore we provide a security and performance analysis for all the proposed protocols.