Scaling is the one of the key stumbling points to blockchain adoption, and has long been the focus of developers trying to push the technology forwards.

Recent events, however, have given Ethereum enthusiasts reason to be optimistic with Ethereum creator Vitalik Buterin explaining in a recent OmiseGO AMA session, that a synergy of first and second layer scaling solutions (Sharding and Plasma) could enable the Ethereum network to manage up to one million transactions per second, enough to "handle most applications" that are currently being developed.

In its current state, the Ethereum network can process a maximum of 15 transactions per second (TPS). A figure proven inadequate by the burden of decentralised apps like CryptoKitties, which tested the limits of the network last December, renewing the focus on developing scaling technologies.

How many TPS does Ethereum need?

VISA, which is often set as a benchmark against blockchain, handles around 2,000 transactions per second, while larger financial institutions like stock exchanges can execute up to eighty thousand transactions a second.

However, this is perhaps a misguided comparison, as Buterin suggests that the majority of the demand for network capacity will come from outside the realm of financial applications, citing IoT devices as potentially creating future demand for up to hundreds of thousands of transactions per second.

More immediate are the ambitious plans entertained by ICOs on the platform, which range from decentralized social networks, to market-driven forecasting technologies, and even complete blockchain jurisdictions.

As a confluence of the demand created by these applications and the capabilities of the platform, an endgame of TPS in the "low millions" is reasonable, Buterin says, although billions is still the "long-term goal."

The route to one million TPS

A hard-coded limit on the number of computations per block caps Ethereum at 15 TPS, which makes scaling solutions necessary for greater network capacity.

Sharding, a Layer 1 scaling solution, is cited by Buterin as the first protocol that will allow Ethereum to scale effectively. Sharding divides the Blockchain network into fragments called shards that contain their own independent piece of state and transaction history, which are then equipped with a group of nodes responsible for processing the information of certain fragments.