It has become evident that as the Ethereum ecosystem has expanded, there are serious limitations on the ability of the Ethereum blockchain to cope effectively and efficiently under high transaction demand.
This comes down to one simple fact: Ethereum, in its current state, does not - and importantly, cannot - scale.
Indeed, Vitalik Buterin (Co-Founder of Ethereum) proclaimed in June 2017 that “Ethereum’s scalability sucks.”
Buterin’s honesty highlights what is currently considered by many to be the largest barrier to Ethereum’s mainstream adoption: the inability to scale beyond 15-25 transactions per second.
There have been numerous ideas proposed to solve on-chain scalability.
The most prominent of these is sharding, which has been laid out as one possible solution, especially to the issue of network congestion and rocketing confirmation times and transaction fees.
Sharding partitions the network into smaller chunks, within which it is faster to reach consensus, as the more manageable chunks are distributed and confirmed by a set number of nodes (rather than the total number of nodes). The chunks are then re-assembled to keep an updated record of the blockchain.
However, given that Sharding is an on-chain solution, it is still unable to prevent itself from falling victim to the same on-chain problems which it attempts to circumvent, most notably blockchain congestion which can result in higher than normal transaction fees.
Another promise is the move from Proof-of-Work (POW) to Proof-of-Stake (POS). PoS however only is another form of leader election/consensus algorithm and will thus not increase the transaction throughput of Ethereum.
Suffice to say that the move to POS will be gradual. Initially only 1/100 blocks will be confirmed via POS while the rest will remain with POW. A complete POS rollout is still years away and not immediately viable.
Ethereum needs urgent solutions to these issues. Otherwise, Ethereum runs the risk of being marginalised by new, faster technologies which are inherently suited to mass use.
Because of Ethereum’s limitations, there has been a radical rethink in our approach to scaling.
The result: Off-chain scaling solutions.
Off-chain solutions offer a way for people, organisations and entities to open up a direct P2P payment channel and execute a transaction without the need to broadcast the transaction across the blockchain.
This has numerous benefits.
Firstly, it makes the cost of the transaction practically free. This can allow transactions of the smallest amounts of currency to be sent across the blockchain enabling a plethora of new uses.
Secondly, transaction speed is vastly boosted by the fact that the transactions are occurring off-chain.
Thirdly, Off-chain transactions don’t add to the traffic on the main blockchain allowing for developers more ‘room’ to experiment with different decentralised applications.
Types of off-chain solutions
2-Party Payment Channels
One of the first exploration of 2-party payment channels was undertaken by Joseph Poon and Thaddeus Dryja in the development of the Lightning Network. This innovative step proposed the possibility of executing transactions off the blockchain through the establishment of payment channels secured with the deposit of collateral. The development of the Lightning Network has led the team behind the Raiden Network to establish the same proposal on top of the Ethereum blockchain. Like Lightning, Raiden requires the collateralisation of assets to secure the trustless nature of 2-party payment channels. Nevertheless, there are still issues with these off-chain solutions which need addressing.
The first issue concerns ‘routing’. In Lightning and Raiden, the process of routing is complex and costly. As users are only able to open direct peer-to-peer payment channels, the transferring of funds to another entity in the absence of a direct connection leads to two imperfect solutions.
First, a user must establish a payment channel directly with the entity. However, the establishment of payment channels is costly and requires collateral to secure. If one has collateralised with one entity, one is unable to use that same collateral with the new entity, as it is already locked in a smart contract.
Second, one can send the funds via a mutual ‘friend’ who has an established connection with both parties. However, once again, this is a costly procedure as, although the entities maybe ‘friends’, the routing via the third party will enable the ‘friend’ to charge a fee for enabling the connection.
These issues will only ever increase the cost of the use of the network. Indeed, Dr Emin Gün Sirer, Associate Professor and Co-Director of the Initiative for Cryptocurrencies and Smart Contracts at Cornell University, has stated that Lighting Network is “economically broken.”
This emanates from the need to collateralise for each single payment channel established on the network. For example, if there are one 1M users of a 2-party payment hub each with $10k worth of transactions, $10B would need to be locked up to reach the required collateral to ensure the trustless nature of the network.
Clearly, with these factors in consideration, the mass adoption of the Lightning Network and Raiden Network is not feasible.
There is however an alternative on the horizon.
N-Party Payment Hubs
The prospect which n-party payment hubs seem to offer indicates that the problems with 2-party payment channels could be overcome. As a result, n-party payment hubs are an attractive avenue to explore and potentially offer the key to the breakthrough in the Ethereum scaling debate.
If it is confirmed that n-party payment hubs can support what they seem to be able to in theory, then the solution for millions of users, transacting across the Ethereum network at a minimal to no-fee, with much reduced collateral and near-instant confirmation time, is ready to be exploited.
This is all made possible thanks to the architecture of n-party payment hubs. The collateral one allocates to the n-party payment hub enables the user to make a payment to any member of the payment hub. This reduces the cost of setting up and maintaining the off-chain payment capability. The movement of funds from one user to the next is achieved with a ‘one time’ deposit and the collateral is managed in bulk.
In such cases, a user could pay another member of the n-party payment hub without the need to collateralise for each payment channel, and in tandem with a mobile app, can set-up and execute a payment instantaneously to potentially millions of users.
This would allow for millions of users to transact Ethereum and all ERC20 based tokens off-chain, dramatically reducing the stress of the main Ethereum blockchain. This would leave the Ethereum blockchain to be used only as a settlement layer to resolve issues arising from disputes.
Given that n-party payment hubs have been deployed for three months at the time of writing, it makes the vision of deploying them across multiple blockchains easy to envisage. If it is possible to deploy n-party payment hubs on one blockchain, then it is entirely feasible to enable them to be built on top of other independent networks.
We will have to wait and see if n-party payment hubs can help blockchains to scale and meet the challenges set by third party payment providers like Visa or PayPal. One thing is certain however, off-chain payment solutions offer a tangible breakthrough which could allow Ethereum to scale. Therefore, the time is right for their implementation.
Dr. Arthur Gervais, co-founder, Liquidity Network
Image Credit: Zapp2Photo / Shutterstock