In our previous article, we thoroughly analyzed the execution layer of ME Network. Today, we’ll shift our focus to the data availability layer, diving deep into its core functions and significance.
What is the Data Availability Layer?
Data availability refers to the process where block producers publish all transaction data of a block to the network, allowing validators to download it. If the block producer publishes complete data and validators can download it, the data is considered available. However, if the block producer withholds some data, preventing validators from accessing the full dataset, that data is deemed unavailable.
In a modular blockchain, the data availability layer plays a critical role in ensuring that network data can be accessed and verified. This typically involves functions like data storage, transmission, and validation to maintain transparency and trust within the blockchain network.
A useful analogy is comparing the data availability layer to the recording of a live football match. Imagine a popular football team playing a match, but not everyone has the chance to watch it live. Therefore, the match is fully recorded and broadcast later. The data availability layer functions like this recording and replay system, ensuring that anyone can watch the match and verify its fairness at any time. Similarly, the data availability layer allows any node to replay the blockchain's history and verify the entirety of its data.
Data Availability is Not the Same as Storage
While data availability involves storing and retrieving data, it differs from traditional data storage. The goal of storage is to ensure that data remains accessible over a long period and continues to have meaning. On the other hand, data availability focuses on the immediate publication of the latest transaction data generated by the main blockchain network (or possibly other networks), making it readily available.
Traditional storage methods are not suited to the needs of the data availability layer. Storage systems typically handle large files, while, for example, the amount of Rollup data needed for data availability between Ethereum blocks is usually less than 2MB. The data availability layer network is designed to handle these small data volumes efficiently through techniques like data sampling for verification and retrieval, which traditional storage networks cannot accomplish on such a large scale.
In essence, data availability is a prerequisite for storage. Only when transaction data is fully published and verified within the network can it support execution, driving state changes and making it worthy of being stored. Storage serves as a pillar for data availability, as the data availability layer generates vast amounts of data, which can be stored on an independent network to relieve the burden on data availability layer nodes and reduce operational overhead.
The Data Availability Layer in ME Network – The Key to Performance
To enhance overall blockchain efficiency, ME Network adopts a layered design, splitting the monolithic chain into an execution layer, settlement layer, and data availability layer. The data availability layer is a critical component for ME Network to achieve high scalability. It handles the reception and storage of execution layer data, acting as a storage module for Rollup data, ensuring transaction data remains accessible and verifiable by anyone. This ensures the security of the execution layer and consensus.
Notably, the data availability layer does not execute any smart contract protocols and is entirely computation-free. Its primary function is to ensure the correct availability of transaction data, providing resources for fraud proofs or, in the event of a network attack, helping users recover assets by referencing the original data and minimizing losses. This means that even without participating in consensus or storing all transaction data, nodes can still verify transactions and determine their availability. The data availability layer is an essential addition to ME Network, offering an additional layer of security for information.
With its unique data distribution, transmission pathways, and storage capabilities, the data availability layer enhances the efficient operation of the main chain (ME Hub), significantly boosting ME Network's overall performance.
How Rollup Data is Accurately Uploaded to ME Network’s Data Availability Layer
This is where data availability sampling (DAS) technology comes into play. DAS allows light nodes to verify the availability of a block without downloading the entire block. Instead, they only sample a portion of the block's data. This is achieved through the use of 2D Reed-Solomon coding (2D RS), ensuring data availability by encoding and sampling, which enables light nodes to validate blocks with minimal trust assumptions.
In ME Network, each block's data is split into smaller sub-blocks and arranged into a matrix. 2D RS erasure coding is applied to these sub-blocks to generate parity data, expanding the matrix. Then, Merkle roots are computed for both rows and columns of this expanded matrix, and these roots are included in the block header as submitted data.
To verify data availability, ME Network's DA light nodes randomly sample sub-blocks from the expanded matrix. Light nodes randomly select unique coordinates and request the corresponding sub-blocks and their Merkle tree proofs from full nodes. If valid responses are received for each sample request, the entire block is likely available.
This sampling process is crucial for the functioning of the data availability layer in ME Network. Without it, coordination between the data availability layer and Rollup would be challenging. In short, a complete data availability layer relies on full nodes that store all data, repeated verification by light nodes, and key technologies like 2D RS erasure coding and Merkle trees.
Conclusion
Whether in traditional monolithic chains or modular chains, the data availability layer plays a pivotal role. As a leader in the modular chain space, ME Network will continue to advance the development and optimization of its data availability layer, exploring more effective blockchain scalability solutions to enhance its network performance.
