A Quick Look at the Rollup Ecosystem: What new designs are there besides Optimistic Rollup and zk Rollup?

Rollup ecology has derived a lot of imaginative design theories and hybrid evolution.

Written by: Alex Beckett Compiled by: Amber

Rollup has come a long way since its conception in early 2019. Now there are two main competitors in the market, and based on this, a lot of imaginative design theories and hybrid evolutions have been derived. This article will take stock of the development status of the Rollup ecosystem as of 2022.

Optimistic Rollup

Since the general-purpose Optimistic Rollup has the first-mover advantage, it has won the majority of Rollup applications, among which Arbitrum and Optimism can also be called "Rollup benchmarks."

Ranked by TVL, Arbitrum currently ranks first in the Rollup app, with a lock-up amount of $2.4 billion. Despite being live on mainnet for over a year, there are still many untested aspects of the system. Offchain Labs, the main body currently developing Arbitrum, operates a single central orderer, and while interactive fraud proofs are flexible, the central orderer is the only whitelisted entity that can submit disputes to generate fraud proofs. From the perspective of users, although the gradual decentralization of the entire system is expected, they still need to give the system a lot of trust.

The next-generation upgrade of the protocol, Nitro, has been launched on Devnet. Nitro will completely change the existing architecture and replace the customized Arbitrum virtual machine with the WASM-Geth combination. Arbitrum's interactive fraud proofs will run on WASM, and nodes will have a Geth-equivalent codebase, plus some Rollup-specific optimizations. All in all, Nitro will bring significant optimizations, providing better EVM compatibility while improving performance.

Optimism is the second largest Optimistic Rollup and ranks third among all Rollups by TVL, with $469 million in locked funds. Optimism PBC operates similarly to Arbitrum, with a single centralized sorter. However, Optimism PBC has found a way to incentivize ecological development by using the profits of the orderers to provide rewards for the network. A total of $1 million in funding from the first round was donated to 58 public goods. Although it is difficult for a centralized design to have a positive impact in the world of "decentralized origin" of the blockchain, the attempts made by Optimism seem to have alleviated this problem to the greatest extent.

Fraud proofs are currently lacking in Optimism's design, although progressive decentralization and necessary security upgrades are to be expected. However, almost all Rollups have immediate or delayed upgradeability, so the security of Rollups ultimately depends on the upgradeability of the multisig design. An upcoming Bedrock upgrade will transform Optimism's architecture to something similar to Arbitrium's. The current Optimism virtual machine will be replaced with a MIPS-Geth combination, where the nodes are equivalent to normal Ethereum Geth nodes. New additions also include Interactive Fraud Proofs, an upgrade to its original non-interactive Fraud Proofs. Bedrock will be another milestone for Optimism, adding many features and improvements to the system.

Additionally, Optimism took a leap forward by experimenting with non-elite governance. Decision-making is divided into two parts, token governance and user governance. Among them, token governance uses tokens as voting certificates, while user governance is a one-person-one-vote system. In such a system, the most difficult point is the distribution of voting identities, which needs to minimize false attacks. If a user can obtain more than one voter status, he can have more influence on the "government" than ordinary users. Optimists point out that they will use non-transferable NFTs to represent citizenship, but the threat of individuals exchanging private keys for more voting power remains.

Fuel is another notable contender in the OptimismRollup space, taking a very different approach from Arbitrum and Optimism. Fuel Labs is building a custom virtual machine for Fuel V2, using the Rust-based programming language. While EVM-compatible rollups are particularly useful for rollups in the Ethereum development ecosystem, custom VMs are easier to maximize performance because they don't have to adhere to many established standards.

The part I like the most about Fuel V2 is the parallel transaction processing capability. Once the amount of data reaches a certain level, the Rollup that implements parallel processing will have more advantages than the Rollup that does not implement parallel processing. Notably, Fuel V1 was the first Optimistic Rollup on Ethereum mainnet, and remains the only Rollup with a permissionless orderer and fraud proofs.

For Fuel, Optimistic Rollup is far superior to zk-Rollups. One is that Optimistic Rollup has complete and unlimited composability: zk-Rollups have an inherent disadvantage in combining smart contracts through zk circuits. The only zk-Rollup that is universally composable is StarkNet, though it currently has a cap on the amount of TVL that permissioned smart contract deployments and bridges can support. All other zk-Rollups are application-specific, or only do token transfers.

Another point is because of EVM compatibility: zk circuits have inherent compatibility issues with certain types of cryptography that are standard in the EVM, making zkEVM an extremely challenging task. Optimistic Rollups have been running on mainnet for some time for EVM compatibility, and the upgrade is expected to further achieve equivalence.

zk-Rollup

For some time now, the Ethereum community has reached a consensus that zk-Rollups are the end state of scalable Rollups. Thus, there appear to be more zk-Rollups in active development than Optimistic Rollups, many of which are scheduled to launch within the next two years.

StarkNet is currently the only general-purpose, composable zk-Rollup on mainnet. However, the system is still in early alpha and has various limitations. There is a limit on the bridge between StarkNet and Ethereum, and the upper limit will gradually increase in the future. Smart contract deployment on StarkNet is also in whitelist mode. I think this is mostly to reduce the risk of smart contract bugs happening, as there likely won't be enough auditors to audit all the contracts a developer wants to deploy - it makes sense for StarkWare to take on this role for the time being. Auditability is one of the common shortcomings of new custom languages, and the complexity of the zk system only makes it more complicated.

zkSync is another major “zk-Rollup” contender, and the recent testnet launch marks the launch of the first instance of zkEVM on a live testnet. However, with the addition of zkPorter, zkSync 2.0 is no longer just an upgrade. It will enable users to choose between zkPorter and Ethereum to publish their transaction data. While ethereum is upgrading data throughput through danksharding, volition is a great middle ground that gives users better options.

While most zk-Rollups prioritize scalability, privacy is another important aspect that zk-Rollups can achieve. Aztec is currently leading the way in the privacy space with their privacy coin transfer zk-Rollup (zk.money), and they are the only privacy-focused Ethereum Rollup that I know of. Aztec is expected to soon launch its next iteration, Aztec connect, which enables users to access Ethereum DeFi privately. This is a big improvement over using an app like Tornado.cash, where privacy can only be achieved by mixing coins and not directly "hiding" transactions.

zk-Rollups are already very complicated, and privacy issues make it even more complicated. It is possible that zk-Rollups will never reach the state of privately composable smart contracts. Because of this, privacy may emerge through specific application chains, whether through zk-Rollups or Validiums on top of zk-Rollup.

Various other zk-Rollups have also emerged, including various projects for Scroll and Polygon. A big difference between zk-Rollups is the difference between using a custom virtual machine or zkEVM execution environment. Its advantages and disadvantages are similar to Optimistic Rollup. However, zk-Rollups have more inherent complexity in implementing zkEVM. Because of this, there is a very strong case for going the route of custom virtual machines and languages like StarkNet and Cairo.

Sovereign Rollup

The last two of the Rollup category are currently only theoretical. A Sovereign Rollup differs from a common Rollup in that it has a fork selection rule that allows it to fork independently of its base layer. Instead, a normal Rollup will delegate its fork choice to its settlement layer, as it needs to ensure the correctness of the Rollup.

Sovereign Rollups are most prominent on DA layers like Celestia, where the correctness of Rollup transactions is not guaranteed by the DA layer. Because of this, Rollup transactions like Celestia are default Sovereign transactions, as they have to ensure their correctness through fraud/validity proofs and fork choice. This should not be mistaken for consensus, the consensus provided by Celestia is an agreement on transaction ordering.

For an optimistic Sovereign Rollup, transactions are considered correct, so Rollup nodes only need to download block data from Celestia. zk-Sovereign Rollup ensures correctness through validity proofs, which will be distributed among Rollup nodes through a p2p network.

The importance of Sovereign Rollups is reflected in their ability to fork, which allows Rollups to be truly independent of their base layer.

Settlement Rollup

The settlement layer is a Sovereign Rollup body built specifically for settlement. Importantly, the settlement layer is any blockchain that has a two-way trust-minimized bridge to Rollup. This bridge enables tokens to be transferred bi-directionally between the Rollup layer and the settlement layer. Trust minimization is a property of bridges where communication relies only on honest minority assumptions proven by verification of data availability and fraud/validity.

Like any settlement layer, the purpose of Settlement Rollup is to provide an environment for "Rollups" to verify proofs, resolve disputes, and bridge tokens. Although, technically, "Rollups" on top of settlement are hybrid in that they use off-chain DA through the data availability layer that the settlement rollup is on, which makes them either a validation layer or an optimistic validation layer.

Hybrid

Validium

Validium is a hybrid zk-Rollup, where transaction data is published off-chain, which means that it relies on the external environment in addition to the settlement layer for validating proofs. StarkEx is the only instance of a validator that has gone live. StarkEx is an application-specific validator that currently supports three applications: Immutable X, Sorare, and DeversiFi. StarkEx also supports zk-Rollup mode, which is used by DyDx, the second largest Rollup application in TVL.

StarkEx utilizes a Data Availability Committee (DAC), which consists of a group of trusted parties to provide data availability for StarkEx validators. While delegating data availability to a permissioned committee reduces security, it enables StarkEx validators to offer cheaper transactions than zk-Rollups. The cost reduction is possible because posting data to Ethereum is expensive, and this is the main variable cost that contributes to Rollup transaction fees.

Some of the security concerns of Validium with DAC can be reduced by using an external data availability layer. The main increase in security comes from the cryptoeconomic security provided by the blockchain, where nodes can be jailed and hacked for dishonest activity. The effectiveness of this implementation is an interesting experiment in the cost-security trade-off of "Rollup".

Optimistic Validity

Like Validium, Optimistic Validium is a hybrid Optimistic Rollup where transaction data is published off-chain. There is no general consensus on what this particular hybrid should be called.

Metis is currently the only Optimistic Validium instance. It is important to note that Optimistic Validium is somewhat weaker than its Validium counterpart, as data availability is required to generate fraud proofs and successfully resolve disputes. If a dispute occurs and data about state transitions is not available, fraud proofs cannot prove fraud. Because of this, if off-chain data availability providers fail to provide data, funds are stolen from Optimistic Validium.

Volition

Through the combination of zk-Rollup and Validium, there is also a hybrid Rollup, namely Volition. Volition allows users to choose data availability on-chain or off-chain. This choice is made at the level of individual transactions, with off-chain data representing cheaper fees and lower security, while on-chain data results in higher fees and greater security. This gives the user the freedom of choice offered by individual systems, rather than explicitly looking for a link that suits the user's cost-safety preferences.

Currently, zkSync 2.0 is the only publicly announced Volition in development. In zkSync 2.0, on-chain data is provided by Ethereum, while off-chain data is provided by their own dedicated PoS chain called zkPorter. However, due to the lack of detailed information disclosure at present, it can only be guessed that StarkWare will provide a user-defined option for StarkEx or StarkNet in the future.

Adamantium

Adamantium is a Validium where everyone personally contributes their own data to the network. An individual's transaction data is stored (off-chain) by the individual, and they must remain online to attest to data availability for each block. If the user is offline, or fails to prove, then their funds are automatically withdrawn to the on-chain settlement layer. While StarkWare has come up with a design for Adamantium, no team has made much progress on it yet. If this assumption can be successfully implemented in the future, it may be very attractive to users or entities who want more personal control over their own security.

Enshrined Rollup

An Enshrined Rollup is a Rollup that consists of a portion of an existing blockchain. Simply put, it is an execution shard. The difference between Enshrined Rollup and Execute Sharding as seen in the Ethereum 2.0 proposal and other similar sharded blockchains is that Enshrined Rollup is proposed to be monolithic. The global validator group will be split into committees and assigned to a specific shard as a validator group. Execution sharding will act as an independent blockchain with its own execution, consensus, and data availability, but it will checkpoint back to the "beacon chain", similar to how sidechains checkpoint back to their chain of choice. In theory, an Ethereum rollup would be execution-only and use the beacon chain to verify data availability and fraud/validity proofs.