The ‘sharding’ is a method by which the processing and storage load is partitioned and distributed throughout a peer-to-peer (P2P) network.
One of the challenges of the blockchain, and other networks based on the same idea of the block chain, is its scalability: expand and expand those distributed networks to address the growing commercial interest without affecting performance. And the rule is simple: as more teams join a network in which all nodes are equal to each other, the effectiveness of the entire system tends to degrade.
This problem of scalability is already being experienced by both bitcoin and Ether (based on the Ethereum network), so it is becoming urgent to find a solution that allows these networks to be extended without limit. That’s where the term we unravel today comes into play: ‘sharding’.
The ‘sharding’ (which we could translate into Spanish as partitioning) is a method by which the processing and storage load is partitioned and distributed throughout a peer-to-peer (P2P) network. In this way, it is achieved that the nodes are not forced to process all the transactional load of the network, but only that related to the corresponding fragment.
Of course, that partitioned information – which can already be seen in traditional databases – can be shared with other nodes, so that the principles of decentralization, immutability and security that characterizes the blockchain are maintained.
The main difference lies in the consensus needed to validate a transaction: in a traditional blockchain, each authentication node registers all the data in the chain and is part of the consensus process. In large block chains, such as bitcoin, most participating nodes must authenticate new transactions and record that information if it is added to the general ledger.
That makes completing each transaction is slow and arduous, saying that the bitcoin network can barely process between 3 and 7 transactions per second. Ethereum, something more advanced, allows to process between 10 and 30 transactions per second. For comparison, conventional platforms that support credit card operators manage more than 1,500 transactions per second.
But with the ‘sharding’ thing changes: every single user account is equivalent to a fragment, and the accounts can only make transactions with other accounts in the same fragment. This allows many parallel transactions to run at the same time, with a separate protocol that allows each of these fragments to be communicated throughout the network.
Who is using it?
Although it is still a concept in diapers within the world of blockchain (with concerns about solving the security of this model), some platforms have already announced that they are exploring their possibilities.
For example, last year, Ethereum began to explore ways to increase the performance of its network, mainly through ‘sharding’.
In turn, seven universities united in the Distributed Technology Research Foundation (DTR) are developing a cryptocurrency network that solves the problems of scalability and blockchain performance through this principle of fragmentation.