Ethereum is moving towards modularity

Concept of Modular Blockchain

Modular blockchains are blockchains that focus on handling a few responsibilities and outsourcing the rest to one or more independent layers. Modular blockchains can be used to handle the following individual or combined tasks:

**Execution: Support the execution of transactions and realize the deployment and interaction with smart contracts. **

**Data Availability: Guarantee the availability of transaction data. **

**Consensus: The content and order of the transactions that are allowed. **

Settlement: Used to complete transactions, resolve disputes, validate proofs, and bridge between different execution layers. **

! [Ethereum is moving towards modularity] (https://cdn-images-1.medium.com/max/1000/0*UkItdv6RdGoqHJnz) modular chains typically perform two or more interdependent functions. For example, the data availability layer must have a consensus on data sorting, otherwise it is impossible to know which data represents the correct version of the history.

Advantages of Modular Blockchain Design

Scalability: Using modularity in a blockchain can increase scale without introducing harmful trust assumptions.

Easy to launch new blockchains: By leveraging modular design, new blockchains can be launched faster without having to worry about keeping every aspect of the architecture right.

Flexibility: Purpose-built modular chains provide more options for trade-offs and design implementations. For example, a modular blockchain system may include a modular chain that focuses on security and data availability, while others focus on execution.

Disadvantages of Modular Blockchain Design

Security: Unlike monolithic chains, modular blockchains do not guarantee the quality of their own security. If the security layers used to handle consensus and data availability are ineffective, modular blockchains are at risk of failure.

Complexity: Implementing a modular blockchain design introduces new complexity. For example, Ethereum's data sharding plan relies on data availability sampling to ensure that nodes on a shard do not hide data. Similarly, the execution layer must create certain complex mechanisms, such as fraud proofs and validity proofs, so that the security layer can guarantee the validity of off-chain state transitions.

Token value: Due to limited applications, the native tokens of some modular blockchains may not be able to absorb value. For example, utility tokens that focus solely on the consensus and data availability layers are used for very little compared to the execution layer, so it may also be more difficult to attract participants to such networks.

Modular form of Ethereum: sharding and rollup

Like first-generation blockchains such as Bitcoin, Ethereum was originally designed as a monolithic blockchain. However, in order to enhance network performance, scalability, and sustainability, the Ethereum network is currently transitioning to a modular framework.

Sharding is the process of splitting a system, such as a database, into parts to run. By distributing functions across multiple components, the system achieves more output and efficiency. In a blockchain network, sharding divides the blockchain into multiple sub-chains, which handle the activities of different parts of the network.

In Ethereum's sharding design, 64 shard chains will run in parallel. Shards can process transactions in parallel (execution sharding) or they can be used to store different parts of the blockchain data (data sharding). With data sharding, Ethereum nodes will only store data published on their shard chain – as opposed to the current structure, which requires all nodes to store the same data.

! [Ethereum is moving towards modularity] (https://cdn-images-1.medium.com/max/1000/0*cK0jaaSCnsFHFvvp) The relationship between Ethereum's beacon chain and shard chain

Sharding is a modular form where different components (shard chains) handle different responsibilities. In data sharding, shard chains store different parts of Ethereum data, and executing sharding allows each shard chain to process its own set of transactions, increasing data throughput and reducing processing time.

Some developers have adopted a rollup-centric approach to scaling Ethereum. Unlike pure off-chain scaling solutions, such as sidechains, rollups are tightly integrated to the main chain. While preserving settlement, consensus, and data availability, the Ethereum blockchain outsources computation to rollups. Since Ethereum acts as the base layer for L2 rollups, rollups can aggressively optimize execution with faster block times and larger blocks without compromising decentralization or security.

! [Ethereum is moving towards modularity] (https://cdn-images-1.medium.com/max/1000/0*CKVe58EXJSowUumz) The functionality of Ethereum (L1 base layer) and rollup (L2) in a modular blockchain architecture

Ethereum's Modular Technology Stack Development Process

The evolution of Ethereum's modular technology stack is as follows:

1. Monolithic blockchain: It represents Ethereum L1 or the main chain, which is itself a monolithic blockchain.

2. Rollup: L2 solutions that act as execution layers, such as Arbitrum and Optimism, move the execution layer out of Ethereum L1, publish state roots and rollup data and pass it back to Ethereum L1.

3. Modular rollups: Rollups with modular data availability.

! [Ethereum is moving towards modularity] (https://cdn-images-1.medium.com/max/1000/1*RKX4YobvmDnOmp0AE9SLSg.jpeg) Ethereum's modular L2 technology stack can provide scalability while retaining a high level of security and decentralization. This powerful combination lays the foundation for Ethereum to become a more efficient and sustainable blockchain ecosystem.

Monolithic blockchain

Monolithic blockchains are the original form of Ethereum, where everything can be processed without the use of rollups or data sharding. This monolithic architecture is the most secure, but comes at the cost of high cost and limited scalability. As a result, the transaction speed of the Ethereum mainnet is relatively slow, with an average TPS of only 15–20. Currently, Ethereum is gradually transforming into a modular blockchain, and this process is accomplished primarily through the adoption of rollup-centric computation and data sharding strategies.

Rollup

Rollup is the earliest technological breakthrough in modular blockchains, extending Ethereum's monolithic architecture by providing a separate layer for execution. Rollups can securely abstract the execution layer of a blockchain to a sequencer, i.e., using a powerful computer to package and execute multiple transactions before periodically passing compressed data back to the Ethereum mainnet for validation. Rollups can increase TPS by 20–50x by moving this computation process off-chain.

In the current scenario, rollups act as the execution layer, processing transactions while outsourcing settlement, consensus, and data availability. For example, optimistic rollups that leverage the Optimistic virtual machine and ZK rollups that run zk EVM. These rollups execute smart contracts and process transactions, but still rely on Ethereum for the following:

Settlement: All rollup transactions are completed on Ethereum. Users of optimistic rollups wait until the challenge period is passed, or after the transaction is deemed valid after fraud prevention calculations have been made. ZK Rollup users need to wait until the validation validity has been proven.

Consensus and data availability: rollups publish transaction data to the Ethereum mainnet in the form of CallData, allowing anyone to execute rollup transactions and rebuild their state if needed. Prior to finality, optimistic rollups require a large amount of block space and a 7–14-day challenge period. Zk rollups store data available for verification for 30 days, providing instant finality, but requiring significant processing power to create proofs.

With Ethereum as the base layer for rollups, rollups can allow for faster block times and larger blocks without compromising decentralization or security. Rollup can be said to be the beginning of a new era for Ethereum. Recently, the total number of transactions between Arbitrum and Optimism has surpassed the number of transactions on Ethereum, reflecting the trend of modularization of Ethereum.

Modular rollups

Newer modular rollups move the data availability layer out of Ethereum. Mantle, for example, still relies on Ethereum's settlement and consensus, but utilizes Mantle DA as a data availability layer. Mantle DA sorts the data and provides proof of the data, but does not need to execute the transaction; Execution transactions are effectively outsourced to Mantle's execution layer.

Previously, Ethereum was the only data availability solution for rollups, resulting in challenges in terms of cost. Data availability is the biggest cost source for most rollups, especially storing transaction data on Ethereum, which can account for up to 70% of the fees. Moreover, this cost is variable, and the cost increases in proportion to usage, gradually becoming a significant barrier as more and more users join. Until now, only large rollups with large resources could accommodate larger user groups.

Thankfully, Ethereum is changing, and new modular solutions are emerging in the form of data availability layers to reduce transaction data submission costs. Key examples of data availability layers include EigenDA, Celestia, and Avail, all of which address data availability issues and provide potential solutions to the limitations of rollups.

The Future of Modularity

Over the past decade or so, the blockchain space has often fallen into a cycle of navigating scalability challenges – constantly creating new L1 blockchains due to Ethereum's high cost and limitations. However, Ethereum's high fees are not an unsolvable bug.

In a world where L2 solutions are becoming the norm for mass adoption, modular blockchains are revolutionizing the architecture of blockchains by dividing layers of execution, settlement, consensus, and data availability. When monolithic blockchains struggle with scalability, the potential of modular architectures is unleashed.

As the data availability layer evolves and competes, the barriers to entry and barriers to entry will be greatly lowered for new rollups. In the not-too-distant future, applications on the OP or ZK stack are likely to boom due to the reduced cost of data availability and the further improvement of modularity.