In-Depth Analysis of Data Availability: Key Challenges and Controversies of L2

Data availability is a crucial concept in the blockchain field. It refers to the process by which block producers publish all transaction data of a block to the network, allowing validators to download and verify it. If the data is complete and downloadable, it is considered available; if hidden data prevents validators from obtaining complete information, it is regarded as unavailable.

It is important to note that data availability and data retrievability are two different concepts. The former focuses on whether the data generated by new blocks can be reached through consensus, while the latter involves the ability to retrieve historical data that has already been stored on the blockchain. Some experts believe that the term "data availability" is misleading and suggest that "data publication" is more appropriate.

In the Layer 2(L2) solution, data availability issues are particularly prominent. The L2 sequencer needs to publish sufficient transaction data for validators to verify transaction validity. However, this process faces two major challenges: first, ensuring the security of the validation mechanism, and second, reducing the cost of data publication.

For Optimistic Rollup, if the sequencer does not publish the complete data, challengers will not be able to initiate valid disputes. In the case of ZK Rollup, although the validity proof itself does not rely on data availability, users still need the complete data to confirm the asset status. Therefore, most L2 solutions currently choose to publish both state data and transaction data on Ethereum to obtain sufficient security guarantees.

However, this approach also brings significant cost pressures. The portion of the Gas fees that users pay to L2, which is allocated to submitting data to L1, constitutes a large part, with the transaction data ensuring data availability being the main component. To reduce costs, the industry has proposed two ideas: one is to lower the cost of publishing data on L1, such as Ethereum's planned EIP-4844 upgrade; the other is to separate the data availability layer from L1, similar to how Rollup separates transaction execution from L1.

This idea of decoupling the data availability layer originates from the concept of modular blockchain. Modular blockchain separates the core functions of the blockchain into four relatively independent parts: execution layer, settlement layer, consensus layer, and data availability layer. By combining different specialized networks, it is possible to break through the performance bottleneck of a single blockchain.

However, there is considerable debate in the industry about whether L2 should separate the data availability layer from Ethereum. Some members of the Ethereum community believe that any scalability solution that does not use Ethereum as the data availability layer cannot be considered a true L2. They are concerned that this could weaken L2's reliance on Ethereum's security, thereby threatening Ethereum's position.

Nevertheless, projects related to the data availability layer are still thriving. With advancements in technology and market demand, we have reason to believe that more innovative data availability solutions will continue to emerge, providing new possibilities for the scalability of blockchain.