How Ethereum 2.0 and Layer 2 solutions can benefit each other

If you have been following the blockchain industry for the past year or so, you will have heard of Ethereum "mergers" or layer 2 solutions. However, these are fairly technical concepts that may be invisible to many people. In this article, the author attempts to provide a more understandable overview of what these things are and how they relate to each other moving forward.

Ethereum’s current structure

As background, it is helpful to understand the current structure of Ethereum and its existing scalability issues. The Ethereum blockchain is a distributed public ledger. Think of it as a large open source operating system (or global computer) where people can build and transact publicly without the need for a centralized authority.

Ethereum Blockchain (and Congestion) Explained

Any transaction that occurs on the Ethereum network is recorded in a "block". For example, if I want to send a token to a counterparty through the Ethereum network, the transaction will be included in a block to be approved before it is sent. Each block is usually composed of many transactions, and these transactions all come from different people. Once a block is accepted by the network, it is linked to the previously accepted blocks (in a data chain), and all transactions within the block can pass. Blocks confirm the exact time and order of transactions and are securely linked together to prevent any block from being changed or a block from being inserted between two existing blocks, so the "blockchain" has tamper-proof and immutable properties.

Things get more complicated when we start exploring the approval process for blocks. Each block is assembled (or mined) by "miners" on the Ethereum network, and people transacting pay them a certain amount of "gas." You can think of gas as a transaction fee or the cost of calculating a transaction.

The size of each block ultimately determines how many transactions can occur on the network at once, and it is determined by the network's block gas limit. As of the time of writing, Ethereum's average block size is 30 million units of gas. In other words, a block is only valid if the total amount of gas consumed by all of its transactions is less than 30 million units. This means that if the network is busy, not all pending transactions can end up in one block (because each transaction requires gas and would exceed the limit), and the network can become congested.

Ethereum Average Gas Limit The chart shows the historical daily average gas limit of the Ethereum network (2016 to present). Unlike Bitcoin, which has a maximum block size of 1 MB, Ethereum technically has no maximum block size (miners can collectively increase or decrease Ethereum's block gas limit). However, limiting the gas consumed per block helps manage the growth of the Ethereum blockchain and the cost of operating a miner or node. Image source: Etherscan

In short, this congestion is an example of why the current Ethereum network is difficult to scale. "Merge" is part of Ethereum 2.0, which aims to increase the speed, efficiency, and scalability of the network, thereby bypassing bottlenecks and processing more transactions at the same time. The current network can process 30 transactions per second, while Ethereum 2.0 promises to process 100,000 transactions per second.

Merger/Ethereum 2.0

The Ethereum 2.0 upgrade is a multi-phase upgrade to the Ethereum blockchain that aims to improve the scalability and security of the network through a number of changes to the network’s infrastructure — most notably, through a switch from a Proof of Work (PoW) consensus mechanism to a Proof of Stake (PoS) model. More details on the differences between the models will be announced soon.

First, the merge is not the only event driving the Ethereum 2.0 upgrade. The upgrade is made up of different parts, and the merge is a part of it (albeit an important part). The three main components are: (i) beacon chain upgrade, (ii) merge, and (iii) sharding. Technically, this involves more upgrades, but these relatively minor upgrades are designed to ensure the smooth operation of the network and will be implemented in parallel with other upgrades.

There are three main components driving the Ethereum 2.0 upgrade.

Each of these is a distinct part of the Ethereum network’s transition to a PoS model. The Beacon Chain was first launched on December 1, 2020, and introduced Proof of Stake to the Ethereum ecosystem. The chain exists as a separate blockchain and currently does not process transactions on the Ethereum Mainnet (short for “Mainnet”). However, the Beacon Chain will soon merge with the Mainnet and become the consensus engine for all Ethereum network data, including execution layer transactions. This official shift to using the Beacon Chain (Proof of Stake) as the block production engine on Ethereum is what the merge represents.

Let’s use an analogy. Imagine Ethereum is a spaceship that is not yet ready for interstellar travel. With the Beacon Chain, the community has built the equivalent of a new engine and a sturdy hull. After significant testing, the old engine is swapped out for the new one for the flight. This will merge the new, more efficient engine with the existing ship, ready to put in some light years and take on the Universe. - Ethereum.org For the merge analogy, the new engine is the Beacon Chain (Proof of Stake), the old engine is Proof of Work, and the spaceship is Ethereum.

The third major component of the Ethereum 2.0 upgrade is shard chains, which will play a key role in scaling the Ethereum network. A common misconception among many is that the Ethereum network will scale and become more efficient immediately after the merger. However, these scaling effects will not begin to take effect until the introduction of sharding, which is expected to happen sometime in 2023.

Sharding essentially splits Ethereum's entire network into smaller pieces, called "shards," in an effort to improve the network's scalability. Instead of centralizing all operations on a single blockchain, shard chains distribute those operations horizontally across 64 new/mini blockchains. By dividing the load, each validator will no longer need to process all transactions on the entire network. This can both reduce network congestion and increase the number of transactions per second on the network compared to the existing proof-of-work model.

Proof of Work (PoW) vs. Proof of Stake (PoS)

Until now, the Ethereum network has been secured through Proof of Work. This model, introduced earlier, includes, among other features, miners for validating blocks. However, after the merger, mining (Proof of Work) will no longer be the means of producing valid blocks on the Ethereum network. Instead, the Proof of Stake validators of the Beacon Chain will assume this role and will be responsible for processing the validity of all transactions.

How does the PoS model differ from the PoW model? In short, instead of having miners verify blocks, Ethereum owners stake their coins in Ethereum’s smart contracts. It is these stakeholders (also known as validators) who are responsible for checking whether new blocks propagated through the network are valid.

Why move to a PoS model? There are several reasons. Chief among them is that the model is more secure for sharding, which, as mentioned above, will play a key role in scaling the Ethereum network.

Sharding breaks the network into multiple parts, and while this structure may increase scalability, it also makes any specific shard in the network more vulnerable to attack. Attacking any specific shard in the network (similar to a mini blockchain) only requires amassing a small portion of the hash rate, otherwise it would require taking control of the entire blockchain without sharding. In addition, with PoW, miners can specifically choose which shards to attack (more strictly speaking, they can choose which shards to contribute their hashing power to). Therefore, PoW miners may collude with each other to focus on a single shard in an attempt to gain control. If miners take over the majority of block producers in a shard, they can manipulate it at the risk of destroying the network. PoS mitigates this risk by not allowing validators to choose which shards they want to work on. Instead, validators are randomly assigned to a single shard, which reduces the chances of collusion between stakers.

“It’s important to note that validators reorganize between shards very frequently (perhaps even once per block), so it’s actually very difficult to target a specific shard for attack. This is a big part of why sharding has been at least theoretically successful in breaking the trilemma.” — Vitalik Buterin in a 2018 speech on how proof-of-stake can help mitigate the security vulnerabilities introduced by sharding.

Compared with the proof-of-work consensus mechanism, the proof-of-stake model is less energy intensive, more decentralized, and more secure.

• Energy Consumption - PoS effectively replaces computing power. It is no longer necessary to use large amounts of energy on proof-of-work calculations to mine.

• Decentralization — The PoS model typically requires more computers and participants (i.e., nodes) on the network to review and approve transactions, which makes it more decentralized than the PoW model.

• Security - In addition to PoS being better protected against 51% attacks through sharding, the model also has a large disincentive for validator attacks. This is largely because validators who ensure the security of the PoS network must stake a large amount of ETH in the protocol (32 ETH as of writing). If these validators attempt to attack the network or behave dishonestly, the protocol can automatically destroy their associated ETH.

Ethereum 2.0 has high goals when it comes to scalability. However, we are still some time away from these upgrades fully taking effect. In the meantime, is there a way to get the Ethereum network to achieve its scalability goals? Let’s discuss Layer 2 solutions next.

Layer 2 Solutions

The main goal of scaling a blockchain is to increase transaction speed (faster finality) and transaction throughput (high number of transactions per second) without sacrificing decentralization or security. It is helpful to think of scaling from two perspectives — on-chain scaling and off-chain scaling.

On-chain Scaling (Layer 1; L1)

On-chain scaling involves changes or upgrades to the Layer 1 network. In the case of Ethereum, this would involve changes to the Ethereum mainnet, which is where transactions are processed and finalized on the network. There are many other Layer 1s, including Solana, Cardano, and Avalanche, each with their own unique characteristics and purposes.

Making changes to Layer 1 can be an extremely complex process and often requires a lot of work. Some changes or upgrades are even close to impossible due to technical limitations. For example, Ethereum's upgrade to proof-of-stake has gone through years of development and is not yet complete.

Given these complexities, off-chain scaling has emerged as another way to scale the network.

Off-chain Scaling (Layer 2)

Unlike on-chain scaling, off-chain scaling does not require any changes to the Layer 1 network. These solutions are built on top of or next to an existing Layer 1 (like Ethereum) and serve as a place where transactions can be made and processed independently of the mainnet. By moving much of the processing burden of Layer 1 to these off-chain networks, the mainnet can focus on other issues such as security without sacrificing scalability. For simplicity, the term "Layer 2" will be used to refer to all off-chain scaling solutions in this article.

Because there are various types of Layer 2, understanding the nuances of each solution can be a challenge.

For example, in the Ethereum context, one of the characteristics of layer 2 is how these solutions derive their security. Some layer 2s, optimistic rollups, zk-rollups, and state channels derive their security directly from the Ethereum mainnet (i.e. layer 1). Within this category, rollups have become the dominant scaling solution, with Arbitrum and Optimism currently garnering nearly $2 billion (both are optimistic rollups).

There are also sidechains, validium, and plasma chains. These layer 2 solutions involve creating new chains whose security is independent of the Ethereum mainnet. Polygon is an example of a popular layer 2 in this category. To date, Polygon has locked nearly $2 billion in total value and supports popular Web3 games and applications such as Zed Run and Quickswap. While not exactly the same, layer 2s can reduce Ethereum's gas fees by 10-100x and process thousands of transactions per second (compared to the current 15-45 transactions per second processed by the Ethereum base layer).

Despite the complexity, having a wide variety of layer 2 solutions helps reduce overall congestion on any one part of the Ethereum network and also prevents single points of failure. Layer 2s can also be optimized for different use cases (DeFi, NFTs, gaming, etc.), so the ecosystem as a whole can benefit from a diverse design.

Can Ethereum 2.0 and Layer 2 Solutions Coexist?

Let’s take a moment to think about the purpose of the Ethereum 2.0 upgrade and the existing layer 2 solutions built on the Ethereum network. At their core, they all aim to solve the same problem - scalability.

Layer 2 solutions are primarily built to address bottlenecks that arise on Ethereum’s base layer. However, once Ethereum 2.0 launches, if the base layer itself is built to scale and becomes exponentially more efficient, then what is the purpose of layer 2? Will layer 2 become obsolete?

To answer this question, it helps to reflect on Ethereum’s vision, which is to make Ethereum mainstream and serve all of humanity. This is a grand vision that needs to scale in the future, considering that more than 4% of the total population owns cryptocurrency. Even if the Ethereum 2.0 upgrade is able to significantly expand the network’s throughput, it remains a question whether it can support the billions of potential daily crypto users in the future world without becoming congested.

Therefore, the relationship between Ethereum 2.0 and its existing layer 2 solutions could be symbiotic. This means that different scalability solutions can work in harmony to have an exponential impact on future transaction speeds and throughput. It is these combined effects, not just one or the other, that Ethereum plans to achieve its vision.

“This is not ‘rollup is not sharding’, but ‘rollup on top of sharding’. That is, rollup is already here or will be here soon, even before sharding, and rollup without sharding can still provide 100x throughput improvement.” — Excerpt from Vitalik Buterin’s Twitter post on Ethereum scaling

Therefore, the massive increase from 15-45 transactions per second (TPS) to 100,000 TPS also takes into account layer 2 solutions in addition to sharding, as these are multiplicative effects. For example, if layer 2 increases throughput by about 100 times, and sharding increases throughput by about 64 times, then layer 2 on top of sharding provides about 6400 times throughput increase. These combined effects may allow the network to scale and meet higher demands in the future.

Others in the Ethereum community have expressed similar thoughts about the future relationship between Ethereum 2.0 and existing Layer 2 solutions.

“As Ethereum L1 becomes more efficient, L2 will also become more efficient while maintaining the current additional benefits.” — Alan Chiu, CEO/Founder of Boba Network

“Even after the merger, in order to truly achieve mainstream adoption, we will need as many scaling solutions as possible.” — Ahmed Al-Balaghi, Co-founder of Biconomy

In the end, the Ethereum 2.0 upgrade will undoubtedly have an impact on the perception and overall utility of layer 2 solutions. However, Ethereum is considering and scaling a future where there is more demand than there is today. If this vision is realized, then layer 2 scaling solutions will continue to play an important role in the ecosystem for a long time.

Original title: "Ethereum: The mutual benefit of Ethereum 2.0 and layer 2 solutions" (Will Ethereum 2.0 Render Existing Layer 2 Solutions Obsolete?)