The number of users of Blockchain applications has increased dramatically. On the Ethereum blockchain, this resulted in the network reaching capability constraints, causing an increase in transaction fees. This is a limiting factor to large-scale adoption. This post highlights the need for scaling solutions and sheds light on some possible approaches.
Recently, I attempted to do a ‘spring-cleaning’ of some old wallets holding the cryptocurrency Ethereum. I am using the word ‘attempted’, because when getting ready to validate the transaction, this is what I saw:
As you can see, the estimated gas fee (transaction fee) is just $9 less than the total amount sitting in the wallet! Realizing how congested the network was, I quickly shelved the project.
This is not an unusual sight. There are currently more than 100,000 active users on Ethereum executing 250,000 transactions per day. In general, transaction fees climb whenever there is an uptick in activity. In 2017, the CryptoKitty craze overtook the network: users could buy, breed and sell virtual cats capable of inheriting traits from their parents. The random inheritance of traits would sometimes result in rare cats which could be sold at a considerable premium.
The frenzy of activity resulted in a staggering increase in network usage, with an associated price hike in transaction fees. Many transactions were not processed for several hours, essentially rendering the network infeasible to use. Since then, scaling Ethereum has turned into one of the most contended topics in the space, gaining more ground in the pandemic Summer of ’20, where DeFi applications took the world by storm, and again during last year’s Bull Run. In both cases, the surge in users and the corresponding increase in average transaction cost is visible on the following chart:
The so-called ‘Blockchain Trilemma’ describes the balance between security, decentralization, and scalability. Scalability in the blockchain world is most frequently referred to as transaction throughput, normally measured in transactions per second. Data and computational scalability are also important when considering the programming of smart contracts. Developers have to consider the size limits for data stored on the blockchain and the number of computations performed, as these increase the cost and time needed to run contracts.It is impossible to optimize for all three criteria, and all blockchains sacrifice one of the three criteria in order to optimize the other two. For example, a system with great security which also has the ability to scale tends to be quite centralized. A system which is decentralized and scalable will tend to have made some calculated tradeoffs regarding security.
In the case of Ethereum, the protocol design favours security and decentralization over scalability. The decentralized design of the network means that Ethereum is capable of handling around 15-23 transactions per second. Of course, if thousands of users want to send ETH around, it will not be long before the network reaches its capability constraints related to the throughput of transactions. The result are gargantuan fees, rendering the network unfeasible to use for the average user as well as developers, thereby putting limits on mass adoption of the digital asset. But surely, there must be a solution?
When talking about blockchain protocols, it is important to distinguish between Layer One and Layer Two. Layer One refers to the actual blockchain, i.e., the protocol layer and all the economic mechanism it entails. Layer Two refers to the technology operating on top Layer one. It is essential to note that Layer Two solutions rely on the security and architecture of the underlying blockchain, and do not attempt to modify the base layer.
In general, there are three main ways to Ethereum or, in fact, most other blockchains.
- Scaling the blockchain itself: Layer One scaling;
- Building on top of Layer One: Layer Two scaling and
- Building on the side of Layer One: sidechains.
Layer One Scaling
As mentioned above, Layer One refers to the actual blockchain, so scaling means modifying the underlying protocol, which could take the form of increasing block size or changing the architecture. The proposed solution for scaling the Ethereum network is a change to the consensus layer, referred to as “Ethereum 2.0” until the end of 2021 before its renaming. This update refers to a set of interconnected changes such changing to Proof-of-Stake (PoS) and sharding. Proof of Stake is a consensus algorithm (a way for the network to reach an agreement) in which a user’s financial resources (their stake) is locked up and ensures correct behaviour. Sharding refers to the process of splitting data into smaller pieces and processing them in parallel on different computers on the network. This can dramatically increase the throughput of the Ethereum network, allowing for about 10,000 transactions per second. Proof of Stake is planned to be implemented in Q2 2022, and sharding between 2022 and 2023.
Layer Two scaling
As for scaling on top of the existing Ethereum blockchain, the most prominent examples of solutions are state channels and rollups.
State channels allow participants to make several transactions off-chain, while only two on-chain transactions are needed (one each to open and close the channel). This type of scaling is ideal if parties know in advance that they will be making a lot of transactions among themselves in the future. This could be the case in a customer-client relationship, or when playing two-person games like chess. As this method decreases the amount of necessary transactions to just two, regardless of how many transactions happen between the two parties, this significantly reduces the transaction cost.
There are already live state channel use cases, for example Perun.
Rollups aim to decongest the network through the batching of transactions. Transactions are compressed and processed off-chain, taking most of the computational load off the network. The results of the computations posted back to the Ethereum Layer One through a transaction. This improves the speed and reduces cost due to the ‘lighter’ transactions. Rollups are considered the most promising scaling solution for the Ethereum blockchain because they allow for scaling of both data and throughput without compromising on security and decentralization. In combination with sharding, data scalability will be achieved as well.
This is not just theory: some rollups are already available for you to try out, for example Arbitrum and Loopring.
Side chain scaling
Side chains are separate blockchains that are compatible with Ethereum. In this scaling approach, developers build on the side of Layer One. There are multiple side chains in existence today, two examples being Polygon and Binance Smart Chain. Due to the fact that these are separate chains employing the Proof of Stake consensus model, they are not constrained by the Ethereum transaction limit, meaning that they can process up to 7000 transactions per second in the case of Polygon, and 58 for Binance Smart Chain.
There is work being put into all three methods, with side-chains and Layer Two scaling solutions already live. It is important to note that different solutions can co-exist and complement each other.
Conclusion | Scale Wars: Ethereum network limitations and how to overcome them
The blockchain industry is ever-changing and dynamic. It is inspiring to see how many teams dedicated themselves to solving the network’s issues. The solutions will contribute significantly to the adoption of blockchain technology, as they are capable of decreasing costs and increase speed of operations on the chain. I am hopeful that there will be user-friendly applications with an intuitive user interface to allow any kind of user to interact with the chains. In short: it is likely that we will see more feasible fees by the next Spring cleaning.
What new capabilities and skills are needed to build a corporate IT that supports and actively drives innovation as part of corporate value creation?
Touted as the greatest revolution since the invention of the Internet, Blockchain will completely reorganize our society. Blockchain as an interesting technical tool holds both opportunities and risks. Only available in German.
Blockchain technology gained enormous market attention in the past years, not only by individuals investing in cryptocurrencies but also by international organizations exploring the potential of a decentral ledger technology deployed to eliminate friction.
Reports on our projects, interesting facts from the various competence and customer areas as well as information about our company have been compiled here for you.