Ethereum is a world Turing-complete state computer that is managed by a network of consensus parties who are actors that verify the whole history of the chain and execute the state of the system to update the ledger's state. Prior to the introduction of layer-2s such as optimistic rollups and the zkEVM, the age-old issue of the scalability trilemma resulted in severely restricted throughput of the underlying layer-1, creating enormous friction to broad adoption. The rivalry for block space only increased with growth, resulting in more expensive transactions that priced out the bulk of network users. To understand the history of layer-2 designs, one must first grasp the scalability trilemma, from which the impetus for layer-2 architectures is readily obvious.

The Scalability Trilemma

The intrinsic scalability restriction of blockchains stems from the fact that each node in the network is needed to store and verify every network transaction. As a result, blockchains are the slowest database on the globe. Blockchains employ peer-to-peer networking, enabling individuals to come and go as they wish while still being able to verify state changes after leaving the network since decentralization has always been a primary principle. However, if a network wished to enhance throughput, it would result in quicker state expansion. Each node would have progressively high computing needs to store the chain's history and maintain its present state. As the computing hardware needed grows, the number of consensus members in the network with the capability to participate diminishes, which is why decentralization and scalability are naturally intertwined. It is not possible to simply increase throughput without losing network decentralization.

Given the fundamental throughput limitations mentioned in the scalability trilemma, what are the best approaches to addressing scalability? 1) Partition the layer-1 state into shards to lower the amount of computing power required for complete chain validation, also known as sharding. 2) The layer-2 strategy, in which transactions are processed on an external computing layer with the same security assurances as if they were done on the underlying layer-1 itself. Layer-2s are best thought of as an information compression method for scalability. Recognizing that not all transactions must be reported on the underlying layer-1, we may instead provide cryptographic proofs of state changes that happened on the external computational layer. As indicated in the rollup-centric vision, the latter approach to scalability has been the major emphasis for Ethereum scalability.

Current rollup technologies

The phrase rollup refers to the process of grouping transactions and giving confirmation of the state change that has happened for that collection of transactions. Instead of authenticating each transaction separately, we combine them into a single solid proof. This is where the information compression comes from since we drastically decrease the number of bytes required for transaction validation on Ethereum's layer-1. A transaction typically consists of 156 bytes, with the digital signature being the most informationally dense portion of the transaction. Layer-2s eliminates the need to personally verify each transaction and instead allows for bulk verification using cryptographic proofs. Layer-2 architectures are classified into two types, which we shall discuss more below.

Optimistic rollups

The word Optimistic rollup is derived from the fundamental economic game that is designed to achieve transaction finality. To elaborate, every transaction in an Optimistic rollup is considered 'optimistically' to be genuine. Instead of giving proof for each state change that happens with each proposed block and its set of transactions, the state transition is believed to be accurate. There is a 7-day contestation period during which actors may contest a state update using fault proofs, which are cryptographic proofs that reveal whether the state transfer was accurate or wrong. If a state transition is not challenged within seven days, the state update is formalized into the optimistic rollups of state history and is regarded as immutable.

optimisitc rollup architecture

Source: Vitalik

In Optimistic rollups, block generation is handled by the sequencer, which fabricates layer-2 blocks (transaction ordering) and conducts the state transition for the group of transactions within the block before finally submitting the relevant data to Ethereum's layer-1 so that data is available to recreate state for the fraud proving system. Optimistic rollups now have a single business operating the sequencer as the parent firm constructing the network, as in Offchain Labs for Arbitrum and Optimism PBC for Optimism. This makes current optimistic rollups very centralized, entrusting block creation to a single organization, although this was required during their guarded launch period. As we leave the guarded launch phase, the sequencer role will become decentralized, letting anybody participate in layer-2 block generation and laying the groundwork for layer-2 tokenomic models.

Zero-Knowledge Rollups

The essential contrast between optimistic and zk rollups is that zk rollups give verifiable evidence of the state transition, while optimistic rollups presume the state transition is genuine unless proven erroneous through fraud-proof resolution. This is where the discrepancy in finality durations arises since zero-knowledge rollups only need the zero-knowledge proof to be confirmed on layer-1, which takes around 45 minutes, but optimistic rollups require the 7-day contention period in order for a transaction to attain finality.

The main issue with zero-knowledge rollups in their present form is EVM equivalence, followed by proof times/costs. The first issue simply implies that layer-2s such as Starkware are not EVM compatible. This implies that native Solidity code representing EVM opcodes cannot be cleanly translated to Starkware and must instead be rewritten in their Cairo programming language. This adds considerable development effort to protocols that seek to make use of the greater scalability provided by SNARKS/STARKS. The ultimate realized goal of zero-knowledge rollups is EVM equivalency. Second, the time and cost to generate validity proofs for state transitions are not entirely optimized, although new zk setups and zk-proof optimized hardware are working hard to solve this issue.

The zkEVM

The zkEVM is the holy grail of scaling solutions, as well as one of the most difficult technical issues in the cryptocurrency sector to date. The purpose of the zkEVM is to completely integrate the Ethereum Virtual Machine (EVM) into an arithmetic circuit used to construct zero-knowledge proofs. Traditionally, in zero-knowledge rollups, a bespoke arithmetic circuit is built for a program so that a zero-knowledge proof may be made to demonstrate that the calculation was run properly. Instead of writing an arithmetic circuit for every single program, which results in severe development overhead, the zkEVM intends to completely implement every single opcode inside the EVM, allowing every smart contract that works within the Ethereum virtual machine to obtain a proof-of-computation without the need for a bespoke arithmetic circuit.

To reduce proof times, the zkEVM uses numerous innovative cryptographic approaches, such as the Poseidon hash function, a zero-knowledge optimized cryptographic hash function, polynomial guarantees, lookup tables, recursive zero-knowledge proofs, and zero-knowledge optimized hardware. Scroll, zkSync, and Polygon are among the projects attempting to construct an EVM-similar arithmetic circuit. In its present form, the codebase for this scaling solution is incredibly extensive, creating a broad attack surface. As a result, the zkEVM will have a lengthy guarded launch period to combat-test the solution. As a result, we anticipate that optimistic rollups such as Arbitrum and Optimism will beat Scroll and zkSync in the short run.

Optimism Bedrock

Since engineers understood that blockchains are built of numerous interconnected systems, each with its own role in the lifecycle of state changes to the ledger, the argument between monolithic and modular design has grown prominent in the area. Modular components conform to the UNIX concept of software development, which divides software into single-purpose modules that do a single job effectively. Engineers may import and reuse these modules, which abstracts a large portion of the labor burden involved with programming and prevents duplicating activities that have already been developed and validated. The same approach applies to blockchains, and the finest example is the in-development Optimism Bedrock Architecture.

Optimism just unveiled its new Bedrock architecture, which proposes that layer-2 rollups may be seen as a modular stack. Modularity has significant advantages in system design and future revisions, as we shall see below. But first, let's go through the layers that make up the modular Bedrock architecture:

Layer of Consensus

The consensus layer serves as the data availability layer, sending current status and payloads to the execution layer. As we shall see later, data availability is critical to the settlement layer, but in brief, in order for the fault-proving system to operate to attest state transitions, the attesters must be able to recreate the state transition using the historical state and payload. Without data availability, you would be unable to prove possibly fraudulent state changes, jeopardizing system security.

Optimism Bedrock replaces the Optimistic Virtual Machine (OVM) that was previously employed in their monolithic layer-2 design with typical Ethereum clients like Geth and Prysm. Separating the consensus layer from the execution layer lays the framework for effortlessly integrating EIP-4844, which depends largely on consensus layer clients, as we shall see later.

Layer of Execution

The execution layer is the virtual environment in which smart contracts are performed, and in the case of Optimism, this is the Ethereum Virtual Machine. This allows smart contracts that have previously been implemented on layer-1 to smoothly migrate to Optimism without requiring any changes to the source. Allowing developers to benefit from scalability without incurring developer expenses.

Because the execution layer is now modular, we can make changes to the execution environment without affecting the consensus or settlement layers. To demonstrate why the preceding is so powerful, consider that we may design an optimistic rollup that does not rely on the EVM but rather on an execution paradigm similar to Bitcoin's UTXO. This essentially suggests that Bitcoin may become an optimistic rollup on Ethereum.

Layer of Settlement

The settlement layer is where transactions reach their finality. This is the fault-proving mechanism that is used to confirm fraudulent state changes in the event of optimistic rollups. The Cannon fault-proving system is a novel fault-proving mechanism introduced by the Bedrock architecture. However, by making the settlement layer modular, the technique utilized to achieve finality may be altered effortlessly. By replacing the fault-proving mechanism with validity proofs, the Bedrock design actually prepares Optimistic rollups to transition into zero-knowledge rollups.

EIP-4844

As previously stated, Ethereum is following a rollup-centric plan in order to achieve scalability while maintaining decentralization. Optimistic rollups are the most production-ready scaling options, having been in use on mainnet since December 2021. Zero-Knowledge rollups, or the zkEVM in general, suffer from non-EVM equivalency or are just now being deployed to mainnet and will undergo the same guarded launch phase with reduced throughput that Optimistic rollups did earlier. As a result, in the near future, it is critical that optimistic rollups minimize the gas cost associated with publishing data to layer 1 in order to boost throughput and cut gas costs for users.

Source: Cryptofee

Optimistic rollups have low costs for layer-2 execution and storage in their present condition, however, in order to comply with data availability. Posting data to layer 1 is still expensive for users. Because data is broadcast to Ethereum's layer-1 through the calldata opcode, both Arbitrum and Optimism have implemented calldata compression methods like Zlib and the brotli compression algorithm. The calldata compression in Optimisms Bedrock architecture really approaches its theoretical limit, implying that the L1 transaction costs have neared their absolute lowest prior to the adoption of EIP-4844.

EIP-4844 is an Ethereum Improvement Proposal that proposes to improve the way data is posted to Ethereum's layer-1 by introducing a new transaction type that is expected to cut layer-2 transaction gas prices by at least 20x. A 'blob-carrying transaction' is the name given to the new transaction type. Instead of publishing data to Ethereum layer-1 through calldata, the blob transaction is saved in the Beacon chain nodes consensus layer and is retained for 14-30 days. Optimistic rollups need data to be accessible for seven days during the contestation phase, but after the transactions are completed, the blob data used for data availability does not need to be maintained and may be erased from the consensus node.

The initial approach to EIP-4844 was to cut the gas cost of the calldata opcode tenfold in order to make data posting to Ethereum's layer-1 significantly cheaper. However, this would raise the average block size, needing more node hardware, and since the data is kept at the execution layer, there is no pruning, forcing nodes to hold this data indefinitely. Proto-danksharding instead saves the blob data in the Beacon chain's consensus layer, which must be retrieved by the consensus clients. This is usually referred to as 'blob space'. Blob space is susceptible to data pruning once the time period for blob data storage for layer-2 data availability has passed. This is much more manageable for consensus clients than permanently keeping the data in the execution layer.

In summary, EIP-4844 is planned to be incorporated in the Shanghai hardfork, which will also permit the withdrawal of staked ETH on the Beacon chain and is projected to significantly cut the cost of layer-2 transactions by at least 20x. We anticipate that EIP-4844 will be the main driving force behind the implementation and eventual broad acceptance of the Ethereum rollup-centric plan. The merging trade for Optimistic rollups is EIP-4844. To explain why we believe this, we shall put out our overarching layer-2 theory below.

The Case for Layer 2

After providing a brief overview of the history of layer-2's, various methods in designs, and the forthcoming EIP-4844 catalyst, the following question will be posed in an effort to formulate a concise thesis that may be used for further research and development:

Using first-principles analysis, users look for low transaction fees given the layer-1 rotation trade that took place in the previous cycle. This trade was driven by Ethereum's fundamental scalability limitation and its refusal to sacrifice the decentralization of the network. Users want low transaction fees because of this. We are of the opinion that the fundamental layer-1 of Ethereum is destined to evolve into the data availability for layer-2's. These layer-2's are the ones that are responsible for the submission of zero-knowledge proofs for zkRollups and the dispute resolutions for optimistic rollups. Because users want minimal transaction costs, and because layer-1 will become a data availability/proof layer, we may conclude that the amount of real on-chain activity on layer-1 is projected to diminish over time and trend toward 0. This conclusion can be reached because of two factors.

Currently, Ethereum's TVL is at $23.95 billion, while Optimism and Arbitrum's TVLs are at $535.7 million and $926.6 million, respectively. It is possible to forecast that over the course of time, money will move from Ethereum layer-1 to some other layer-2 network due to the fact that users want cheap transaction costs and liquidity wants users. It is anticipated that the implementation of EIP-4844, a key catalyst for the layer-2 thesis, would result in a reduction in transaction costs on optimistic rollups of at least 20 times. Following the activation of EIP-4844, we anticipate a rapid increase in the flow of money into optimistic rollups. Within the next five years, we anticipate that around seventy percent of Ethereum's layer-1 TVL will transition to a layer-2 architecture from a longer-period perspective. As the process of onboarding new users from CEXs or other parties evolves, we anticipate that layer-2 networks will become the standard network for onboarding new users and that new users will interface with Ethereum layer-1 very seldom in the future.

After the merge, one of the most important questions to ask when constructing a portfolio is, "What assets outperform Ethereum?" Given the decrease in ETH issuance of almost 90 percent and the fact that times of high network demand result in ETH being deflationary as a consequence of the burning of the base fee in EIP-1559, this is really a question that is impossible to answer. Regarding other layer-1s, we presently believe that their token economies are worse since they have bigger issuances for validators, which results in a greater quantity being released into the market each year. We have the impression that attention is currently shifting back to Ethereum following the layer-1 rotation space and ultimately to the rollup-centric roadmap that Ethereum has developed, which is currently in the process of being fully realized and is prepared for the subsequent phase of mass adoption. According to the first principles reasoning shown above, we believe that layer-2s will behave as a strong beta on ETH and have the highest chance of outperforming ETH on a yearly basis.

The Prospects for Layer-2's in the Future

The Sequencer Will Be Decentralized.

At this current juncture in history, the layer-2 sequencers are consolidated into the entities that were responsible for the network's development. Offchain labs, representing Arbitrum, and Optimism PBC, representing Optimism, are the entities in question. They are the only people who can operate the sequencers, which contributes to the extremely centralized nature of layer-2's in their current configuration. However, gradual decentralization is strongly advised for new computational layers that have high attack surfaces in their initial guarded launch phase. This is because these new computing layers are more vulnerable to attacks.

A proof-of-stake mechanism, in which users would be allowed to stake the native L2 token and which imposes slashing conditions for malicious behavior in order to disincentivize the posting of invalid state transitions, which can always be affirmed and proved incorrect using the Cannon fault proving system, would most likely be the means by which the sequencer would be decentralized. From this vantage point, we are able to hazard a guess on likely layer-2 tokenomics resulting from decentralizing the sequencer:

Transaction Fees: Those individuals who stake their tokens and take part in the sequencer are eligible to get a percentage of the transaction fees that are collected on layer 2. This is a clear incentive for progressively decentralizing the job of the sequencer.
MEV: Transactions may be reordered during the building of layer-2 blocks in order to offer MEV services, and a part of the income generated by these services can be given to the block producer.
Staking Rewards: Those that take part in the sequencer role will be awarded the L2 native token as a reward for their honest behavior as a result of staking. Those that stake their tokens and take part in the construction of blocks are eligible to get a reward from the token inflation.

MEV

There have been discussions regarding the possibility of eradicating harmful forms of MEV, such as front-running and sandwich attacks, through the implementation of a timelock-encrypted transaction mechanism. This mechanism would conceal transactions until they were validated and added to the proposed layer-2 block. The following is how the timelock encryption technique operates:

The user generates a key by solving a time-lock challenge in order to encrypt the transaction.
The time-lock puzzle, the tx, and the validity evidence are all sent to the sequencer.
The sequencer checks the validity evidence and incorporates an encrypted tx on a first come, first served basis.
The sequencer decrypts the transaction after successfully resolving the timelock problem using the key

To put it more simply, the sequencer is unable to see the content of the transaction, and as a result, it is unable to reorganize transactions inside a block in order to execute MEV and obtain value from users of the network. The latency of confirmation times will be longer as a result of this change, however, malicious MEV such as frontrunning and sandwich attacks, will no longer be possible. zkSync has already communicated its desire to learn more about this MEV preventive strategy.

Appchains at the Layer 3 Level

The modularity of Layer 2 makes the implementation of appchains possible. The 'opchain' sequencer that was included in the first Optimism layer-2 may now be inherited by individual projects, allowing such projects to install their very own 'opchain.' The term "superchain" refers to the network of linked opchains that may eventually be created as a combination of many distinct projects, each creating its own opchain. Because they will be using the original layer-2 sequencer, opchains will be able to conduct entirely cross-chain atomic transactions. This is something that is absolutely necessary for opchains to be able to work together. The basic idea behind this architectural style is that opchains and appchains, in general, do not want the performance of their networks to be negatively impacted by other decentralized applications (DApps), which is something that regularly occurs on Ethereum's monolithic layer-1. As a result, the application-specific rollups or sovereign chains that Cosmos has suggested do, in fact, have a role in the future.

In addition, zkSync is working on the development of "HyperChains," which are connected blockchains that are both customizable and trustless and are driven by zkSync's layer-2 zkEVM scaling solution. The developers of a chain have the ability to make customizations while still inheriting the security features of Ethereum. Options like on-chain privacy and matching tokenomics designs with the chain itself may be customized, as can the data availability solutions that power the HyperChain. Customization can also take place in the data storage solutions that fuel

HyperChain. "HyperBridges" are what link each of these layer-3 appchains to one another. As we can see beyond the Cosmos environment, which aspires to allow a full ecology of linked sovereign chains, the Ethereum layer-2 scalability solutions are also headed in the same way. This is something that we can see outside the Cosmos ecosystem.

Source: zk Sync

Conclusion

The scaling plan of Ethereum places a significant emphasis on Layer-2s. The layer-1 rotation trade occurred because Ethereum was unable to scale in the allotted amount of time to allow for the growing number of users. Optimistic rollups have now left the cautious launch phase of their development and are carrying on with the process of lowering transaction costs by using call data compression methods. The final trigger that will bring about the instant scalability of optimistic rollups is EIP-4844, and it will be implemented soon. Given the large drop in layer-2 transaction costs that is predicted to take place, we may anticipate a migration of users and liquidity to both Optimism and Arbitrum.

Ethereum is, without a doubt, the most widely used smart contract blockchain, with the most extensive developer community as well as the most advanced development tools. It is reasonable to anticipate that these developers would move their already existing DApps or freshly produced DApps to an optimistic rollup in order to lower the cost of gas for their consumers. We anticipate that the majority of on-chain activity will take place on optimistic rollups over the next three years, and as a result, we are in an excellent position to capture an enormous amount of value as liquidity and on-chain activity move away from the underlying layer-1 in search of a cheaper transaction environment

Disclaimer:
This article is intended for and only to be used for reference purposes only. No such information provided constitutes advice or a recommendation that any investment or trading strategy is suitable for any specific name.

Rollup techs and EIP 4844: The catalysts for Layer 2