Scalability is by far the most pressing issue yet to be addressed by blockchain. Without being able to accommodate the transaction loads imposed by its users, a blockchain is unfit for purpose.
What’s the problem
The two main blockchains, Bitcoin, launched in 2009, and Ethereum, launched in 2015, are crippled by a lack of scalability. Bitcoin can handle around seven transactions per second (tps), while Ethereum tops out at approximately twice that number. Per second. Compare this to Visa, an example of an effective and widely-adopted network, which is comfortably capable of handling loads of up to 24,000tps and a typical load of around 6,000tps. To truly act as a replacement for a network that can handle enterprise-scale transaction throughput, it’s immediately obvious that Bitcoin and Ethereum, and every other blockchain languishing at transaction capacities below 5,000tps, have some serious work to do.
It must be mentioned, however, that not all blockchains need to scale. Bitcoin can happily exist as a “store of wealth” without needing massive scalability. Banks successfully keep gold in their vaults as a physical store of wealth without the need for rapid access and high throughput, and Bitcoin can perform the same role for digital wealth.
Scalability is essential
Yet similar to gold, Bitcoin fails as an everyday currency in that it’s impractical to use when buying groceries; the transactions are too slow — often taking many hours to confirm — and the network cannot handle the load when flooded with even a modest number of transactions.
Other specialist blockchains, such as Monero, also don’t necessarily need to prioritize scaling. Monero’s primary focus is on ultra-private transactions, and the overwhelming majority of blockchain users don’t require that level of privacy on a regular basis.
Expressed differently, hiding the fact that you bought a box of icing sugar and a bunch of mixed herbs at the supermarket isn’t a priority, while hiding the fact that you bought five grams of Columbia’s finest and a quarter pound of weed from an online drug dealer is.
As such, Monero can get away with passing the buck when it comes to scaling, as it’s capable of handling the specialized types of transactions and transaction loads for which it was designed.
In particular, Ethereum’s shortcomings in the scalability department have brought the issue to the forefront of blockchain development. During 2017, at the height of the ICO boom, the Ethereum network was frequently brought to a virtual standstill by thousands of simultaneous transactions, resulting not only in frustrated investors, but also frustrated users whose unrelated activities were hindered by the congestion. Similarly, the popular “CryptoKitties” game recently brought the Ethereum network to its knees, with tens of thousands of transactions queued and delays of up to twelve hours. Clearly, this problem has to be solved before blockchain can progress to the next level.
To their credit, efforts are being made by Bitcoin and Ethereum to address the scalability issue, with projects such as the Bitcoin’s Lightning Network, and Ethereum’s Raiden and Plasma. These solutions, however, are “bolt-on” efforts to patch over a fundamental issue with the underlying blockchain. Furthermore, internal politics and bureaucracy within the Bitcoin and Ethereum communities makes progress towards adopting a solution extremely slow.
What is sharding?
The practical solution to scalability is simple: build a scalable blockchain correctly from the ground up. This is precisely the path being taken by Emotiq, which is creating a blockchain with scalability embedded into its DNA by adopting an elegant solution known as sharding. But what is sharding, and how will it solve the scalability problem?
To illustrate the concept of sharding in a simple manner, picture a blockchain as a network of nodes, all of which are involved in verifying transactions. And picture, say, a network consisting of 6,000 nodes, roughly the number in the Bitcoin network.
In a typical blockchain, all of these nodes will be working on verifying the same transactions at the same time. This is what makes blockchain technology so robust, because small numbers of malicious nodes are kept in check by the vast majority of honest nodes. Yet to have all the nodes working on the same transactions at the same time is inefficient. And let’s assume, also for the sake of simplicity, that the network in this example can process 10tps, regardless of the number of nodes present, as they’re all working in unison and doing exactly the same work as each other. More nodes does not equal greater capacity.
Sharding is method of increasing transaction capacity by splitting a large network into independent groups (shards) of nodes, with each shard working on different transactions rather than the entire network processing the sametransactions. In our example, the 6,000 node network could be sharded into ten separate shards, each containing 600 nodes (a sufficient number to guarantee integrity), and each processing transactions at a rate of 10tps. One doesn’t need a Ph.D. in mathematics to work out that the total throughput of our theoretical sharded network is 100tps (ten shards operating in parallel at 10tps), ten times greater than the unsharded network, and — critically — without any meaningful loss of decentralization or integrity. Furthermore, if the network size doubles to 12,000 nodes, the unsharded version can still only process 10tps, while the unsharded network can now process 200tps, twenty times more transactions per second. The larger the network, the greater the benefits of parallel transaction processing through sharding.
Emotiq and sharding
When a sharded network is coupled with other blockchain innovations ,such as rapid Proof-of-Stake (PoS) consensus, transaction capacity increases dramatically. Emotiq is aiming for an initial target of 5,000tps based upon PoS consensus and sharding in a network consisting of just 1,800 nodes. From there, the sky’s the limit.
Sharding is not a blockchain architecture that is easily added as an afterthought. To create a network with strong, uncompromised scalability, sharding needs to be there at the beginning — which we’ve achieved with Emotiq.
Taking a ground-up approach also yields another benefit: the opportunity to integrate other hard-to-add features into the core blockchain, such as privacy and accessibility. Solving the scalability problem alone will not miraculously produce wide-scale adoption. It takes more than one pillar to hold up a structure.
Solving the scalability, privacy, and accessibility problems in one single blockchain is the Holy Grail, and the quest Emotiq has undertaken.
➤ Join our Telegram community
➤ Follow us on Twitter
➤ Visit our website
