Addressing Scalability and Storage issues in Block Chain using Sharding

Abstract

There is no skepticism about the perpetuity of recorded transactions in BlockChain and its Distributed Ledger Technology. But the hype is casting shadow over the real world implementation issues. Issues like limited scalability and storage constraints impede mass adoption of this world-changing technology. In this paper, we will deliberate on the issues like scalability and storage in BlockChain Technology. We will provide solutions to overcome these problems using Sharding along with distributed load balancing algorithms. The requirement of nodes increases as the number of transactions increases in a network. Preserving the essence of BlockChain Technology, if we can cleave whole system memory into local and global Shards along with dedicated miners at local shards, we can cater to needs of scalability and storage. Distributed load balancing algorithms will be implemented in network to divide the transactions recorded in distributed public ledger among different local shards and global shards. Equal distribution of transactions at local shards will increase the resource utilization of the network. Dedicated miners on these local shards will validate the transactions, hence decreasing the network latency and overhead. Distributed Proof of Stake (PoS) Consensus protocols will validate the transactions by aggregating the results of local and global shards and will update the copy of the public ledger. To meet scalability demands, the system will expand to global shards with global miners assigned dynamically. And partitioning the ledger into local and global shards to be copied to different nodes in the network will address the storage challenges in BlockChain technology. The main contributions of this paper are: 1) Using distributed load balancing algorithms to copy local shards of distributed public ledger to different nodes dynamically with replication 2) On demand expansion of nodes from local to global shards to cater scalability 3) Distributed PoS consensus protocol executed on aggregated results to update the public ledger 4) Performance of the system will improve as load distribution of shards across the nodes and aggregation of results will work in parallel.