In-depth Analysis of Ethereum Scaling "Holy Grail" Rollups: Transactions Could Rise to Thousands Per Second, Potentially Triggering Another High Growth Period for DeFi?

Original author: jakub, Finematics

Compiled by: Glendon, Babitt

Rollups are considered the ultimate solution for scaling Ethereum. So, what are rollups? What is the difference between Optimism and ZK rollups? How does Arbitrum differ from Optimism? And why are rollups considered the holy grail when it comes to scaling Ethereum? You will find answers to these questions in this article.

## 1. Introduction

Scaling Ethereum has been one of the most discussed topics in the crypto space. The scaling debate often heats up during periods of high network activity, such as the CryptoKitties craze in 2017, the DeFi summer of 2020, or the cryptocurrency bull market in early 2021.

During these times, the unprecedented demand for the Ethereum network led to extremely high gas fees, making it expensive for everyday users to pay transaction fees.

To address this issue, finding the ultimate scaling solution has been one of the top priorities for multiple teams and the entire Ethereum community.

Generally, there are three main ways to scale Ethereum or most other blockchains: scaling the blockchain itself—Layer 1 (L1) scaling; building on top of Layer 1—Layer 2 (L2) scaling; and building sidechains alongside Layer 1.

For Layer 1, Eth2 is the preferred solution for scaling the Ethereum blockchain. Eth2 refers to a set of interconnected changes, such as migrating to Proof of Stake (PoS), merging the state of the Proof of Work (PoW) blockchain into the new PoS chain, and sharding.

Specifically, sharding can significantly increase the throughput of the Ethereum network, especially when combined with rollups.

When it comes to scaling beyond Layer 1, various scaling solutions have been attempted with mixed results.

On one hand, we have Layer 2 solutions like Channels, which are fully secured by Ethereum but only applicable to specific applications.

On the other hand, sidechains are often EVM-compatible and can scale general applications. The main drawback is that they do not rely on Ethereum's security but have their own consensus models, making them less secure than Layer 2 solutions.

Most rollups aim to achieve the best of both worlds by creating a general scaling solution that still fully relies on Ethereum's security.

This is the holy grail of scaling because it allows all existing smart contracts to be deployed on Ethereum with minimal changes, without sacrificing security.

It's no wonder that rollups may be the most anticipated scaling solution of all.

But first, what are rollups?

## 2. Rollups

A rollup is a scaling solution that works by executing transactions outside of Layer 1 while posting transaction data to Layer 1. This allows rollups to scale the network while still deriving their security from Ethereum's consensus.

Moving computation off-chain effectively allows for more transactions to be processed since only a portion of the rollup transaction data needs to fit into an Ethereum block.

To achieve this, rollup transactions are executed on a separate chain, which can even run a rollup-specific version of the EVM.

The next step after executing rollup transactions is to batch them together and publish them to the Ethereum main chain.

The entire process essentially involves executing transactions, obtaining data, compressing the data, and aggregating it into a single batch for the main chain, hence the name "rollup."

While this seems like a potentially good solution, a natural question arises:

"How does Ethereum know that the published data is valid and not submitted by malicious actors trying to benefit themselves?"

The exact answer depends on the specific rollup implementation, but generally, each rollup deploys a set of smart contracts on Layer 1 responsible for handling deposits and withdrawals as well as validating proofs.

Proofs are also where the main differences between the types of rollups lie.

Optimistic rollups use fraud proofs. In contrast, ZK rollups use validity proofs.

Let’s delve deeper into these two types of rollups.

## 3. Comparison of Optimistic and ZK Rollups

Optimistic rollups publish data to Layer 1 and assume it is correct, hence the term "optimistic." If the published data is valid, we are on a happy path and do not need to do anything else. Optimistic rollups benefit from not needing to do any extra work under optimistic assumptions.

In the case of invalid transactions, the system must be able to identify them, recover the correct state, and penalize the party submitting such transactions. To achieve this, Optimistic rollups implement a dispute resolution system capable of verifying fraud proofs, detecting fraudulent transactions, and deterring malicious actors from submitting other invalid transactions or incorrect fraud proofs.

In most implementations of Optimistic rollups, the party able to submit batched transactions to Layer 1 must provide a bond, typically in the form of ETH. Any other network participant who discovers an erroneous transaction can submit a fraud proof.

After submitting a fraud proof, the system enters a dispute resolution mode. In this mode, the suspicious transaction is re-executed on the Ethereum main chain. If the executed proof transaction is indeed fraudulent, the party submitting this transaction will be penalized, usually by having their staked ETH slashed.

To prevent malicious actors from spamming the network with incorrect fraud proofs, parties wishing to submit fraud proofs typically must also provide a bond that may be subject to slashing.

To be able to execute rollup transactions on Layer 1, Optimistic rollups must implement a system capable of replaying transactions in the exact state they were executed on the rollup. This is one of the complexities of Optimistic rollups, often achieved by creating a separate manager contract that replaces certain function calls with the rollup's state.

It is worth noting that even if only one honest party monitors the rollup's state and submits fraud proofs when necessary, the system can work as intended and detect fraud.

It is also worth mentioning that, due to the correct incentive mechanisms within the rollup system, entering the dispute resolution process should be a rare occurrence rather than a frequent event.

When it comes to ZK rollups, there is fundamentally no dispute resolution process. This is achieved by leveraging a clever cryptographic technique called zero-knowledge proofs, hence the name ZK rollups. In this model, each batch published to Layer 1 contains a cryptographic proof known as a ZK-SNARK. This proof can be quickly verified by Layer 1 contracts when submitting the transaction batch and can immediately reject invalid batches.

Sounds simple, right? Perhaps on the surface. In practice, countless hours have been spent by multiple researchers iterating on these clever cryptographic and mathematical works to make it work.

There are also some other differences between Optimistic and ZK rollups, so let’s go through them one by one.

Due to the nature of the dispute resolution process, Optimistic rollups must give all network participants enough time to submit fraud proofs before finalizing transactions on Layer 1. This time period is usually quite long to ensure that even in the worst-case scenario, fraudulent transactions can still be contested.

This leads to a relatively long time to withdraw funds from Optimistic rollups, as users must wait even up to one or two weeks to withdraw their funds back to Layer 1.

Fortunately, some projects are working to improve this situation by providing fast "liquidity presence." These projects offer nearly instant withdrawals back to Layer 1, another Layer 2, or even a sidechain, charging a small fee for this convenience. Hop Protocol and Connext are projects worth watching.

ZK rollups do not have the long withdrawal times issue because funds can be withdrawn as soon as the rollup batch is submitted to Layer 1 along with the validity proof.

So far, it seems that ZK rollups are just a better version of Optimistic rollups, but they also have some drawbacks.

Due to the complexity of the technology, creating an EVM-compatible ZK rollup is much more challenging, making it harder to scale general applications without rewriting application logic. That said, ZKSync is making significant progress in this area, and they may soon launch an EVM-compatible ZK rollup.

Optimistic rollup implementations are somewhat easier in terms of EVM compatibility. They still need to run their own version of the EVM with some modifications, but 99% of contracts can be ported without any changes.

ZK rollups also have a higher computational load than Optimistic rollups. This means that nodes computing ZK proofs must be high-spec machines, making it difficult for other users to run them.

In terms of scaling improvements, both types of rollups should be able to scale Ethereum from 15 to 45 transactions per second (depending on the type of transaction) to 1,000 to 4,000 transactions per second.

It is worth noting that by providing more space for rollup batches on Layer 1, more transactions can be processed per second. This is also why Eth2 can create significant synergies with rollups, as it increases the possible data availability space by creating multiple shards—each shard capable of storing a large amount of data. The combination of Eth2 and rollups could potentially increase Ethereum's transaction speed to 100,000 transactions per second.

Now, let’s talk about all the different projects working on Optimistic and ZK rollups.

## 1. Optimistic Rollups

Optimism and Arbitrum are currently the most popular projects in the Optimistic Rollups space.

Optimism has partially rolled out on the Ethereum mainnet, working with limited partners like Synthetix or Uniswap to ensure that the technology works as intended before a full launch.

Arbitrum has deployed its version on the mainnet and started incorporating different projects into its ecosystem. Instead of only allowing a limited number of partners to deploy their protocols first, they decided to provide a time window for all protocols wishing to launch on their rollup. After this time window ends, they will open it up to all users at once.

Some of the most notable projects launched on Arbitrum include Uniswap, Sushi, Bancor, Augur, Chainlink, Aave, and more.

Recently, Arbitrum also announced a partnership with Reddit. They will focus on launching a separate rollup chain that will allow Reddit to scale their rewards system.

Optimism is working with MakerDAO to create the Optimism Dai bridge, enabling DAI and other tokens to be quickly withdrawn back to Layer 1.

Although Arbitrum and Optimism are both trying to achieve the same goal—building an EVM-compatible Optimistic Rollups solution—there are some differences in their designs.

Arbitrum has a different dispute resolution model. They do not re-run the entire transaction on Layer 1 to verify the validity of the fraud proof; instead, they propose an interactive multi-round model that allows narrowing down the scope of the dispute and possibly executing only some instructions on Layer 1 to check the validity of suspicious transactions.

This also has the nice side effect that smart contracts deployed on Arbitrum can potentially be larger than the maximum contract size allowed on Ethereum.

Another major difference is how they handle transaction ordering and MEV.

Arbitrum initially will run a sequencer responsible for transaction ordering, but in the long run, they hope to decentralize it.

Optimism prefers another approach, where the order of transactions and MEV can be auctioned to other parties over a certain period.

There are also several other projects dedicated to Optimistic Rollups, such as Fuel, the OMG team with OMGX, and Cartesi, among others. Most of them are also trying to develop an EVM-compatible version for their rollups.

2. ZK Rollups

While it seems that the Ethereum community is primarily focused on Optimistic Rollups, at least in the short term, we should not forget that projects dedicated to ZK rollups are also making rapid progress.

For ZK rollups, we have several options available.

Loopring uses ZK rollup technology to scale its trading and payment protocols.

Hermez and ZKTube are expanding payment scalability using ZK rollups, with Hermez also building an EVM-compatible ZK rollup.

Aztec is focusing on introducing privacy features into its ZK rollup technology.

Projects like DeversiFi, Immutable X, and dYdX have widely adopted StarkWare-based rollup technology.

As we mentioned earlier, ZKSync is developing an EVM-compatible virtual machine that will be able to fully support any smart contract written in Solidity.

## 4. Conclusion

As we can see, there is a lot going on in both the optimistic and ZK rollup camps, and the competition between different rollups will be interesting.

Rollups should also have a significant impact on DeFi. Users who previously could not transact on Ethereum due to high transaction fees will be able to stay in the ecosystem during the next period of high network activity. They will also support new types of applications that require cheaper transactions and faster confirmation times. All of this is fully protected by Ethereum's consensus. It seems that rollups could trigger another high-growth period for DeFi.

However, there are still some challenges regarding rollups.

Composability is one of them. To compose transactions using multiple protocols, all of these protocols must be deployed on the same rollup.

Another challenge is liquidity fragmentation. For example, if no new funds enter the Ethereum ecosystem as a whole, the existing liquidity in Layer 1 protocols like Uniswap or Aave will be shared between Layer 1 and multiple rollup implementations. Lower liquidity typically means higher slippage and poorer trade execution.

This also means that there will naturally be winners and losers. Currently, the existing Ethereum ecosystem is not sufficient to take advantage of all scaling solutions. This is likely to change in the long run, but in the short term, we may see some rollups and other scaling solutions become deserted.

In the future, we may also see users living entirely within a rollup ecosystem and not interacting with the Ethereum main chain and other scaling solutions for extended periods. This could be especially evident if we see more centralized exchanges able to facilitate direct deposits and withdrawals.

Nevertheless, rollups seem to be the ultimate strategy for scaling Ethereum, and the challenges are likely to be mitigated in some way. It is certainly interesting to see how rollups gain increasing user adoption.

One question that often arises when discussing rollups is whether they pose a threat to sidechains. Personally, I believe that sidechains will still have a place in the Ethereum ecosystem. This is because, although transaction costs on Layer 2 will be much lower than on Layer 1, they are likely still high enough to price out certain types of applications (like games and other high-capacity applications).

This situation may change when Ethereum introduces sharding, but by then, sidechains may have generated enough network effects to survive in the long term. It will be very interesting to see how this develops in the future.

Additionally, the fees for rollups are higher than for sidechains since each rollup batch still needs to pay for Ethereum block space.

It is worth remembering that the Ethereum community is very focused on rollups in the Ethereum scaling strategy—at least in the short to medium term, and possibly even longer. I recommend reading Vitalik Buterin's article on a rollup-centric Ethereum roadmap.