Understanding the 'DA Competition' that will determine Ethereum's future in one article: Who will be the ultimate winner among Celestia, EigenDA, and Avail?

2024-01-29 12:01

Read this article in 17 Minutes

Avail, EigenDA, and Celestia are the main players in the DA ecosystem - each serving the same space but taking slightly different approaches in terms of infrastructure stack, execution, and listing.

The data availability layer has become an important component of modular architecture, serving as a plug-in component to reduce costs and expand the blockchain. The core function of the DA layer is to ensure that on-chain data is available for use and access by all network participants. Historically, each node had to download all transaction data to verify its availability, which was an inefficient and costly task. This is how most blockchains currently operate and is a barrier to scalability, as the amount of data required for verification increases linearly with block size. Ultimately, end-users suffer here: data availability costs account for 90% of the transaction costs incurred by users trading on Rollup (currently, the cost of sending transaction data to Ethereum on Rollup is $1300-1600/mb).

The introduction of Data Availability Sampling (DAS) fundamentally changes this architecture. With DAS, light nodes can confirm data availability by participating in multiple rounds of random sampling of block data, without having to download each entire block. Once multiple rounds of sampling are completed and a certain confidence threshold for data availability is reached, the rest of the transaction process can proceed safely. In this way, the chain can expand its block size while maintaining simple data availability verification. Additionally, significant cost savings are achieved: these emerging layers can reduce DA costs by up to 99%.

A very appropriate analogy of DA in 0xngmi

除了实现更高的吞吐量之外，数据可用性层对于提高互操作性也很有意义。廉价的 DA 将不可避免地推动新的定制 Rollup 链的寒武纪爆发，通过 Caldera、AltLayer 和 Conduit 等 Rollup-as-a-service 提供商使部署变得越来越简单。然而，随着 L2 和 L3 生态系统的出现，它们默认会变得支离破碎。让用户使用新平台已经很困难了——如果互操作性、流动性和网络效应有限，情况会变得更糟。统一的 DA 层作为每个网络的基础，会使得资金流变得更加简单，并吸引更广泛的用户群。

Translation:

In addition to achieving higher throughput, the data availability layer is also meaningful for improving interoperability. Inevitably, cheap DAs will drive the explosion of new customized Rollup chains, making deployment increasingly simple through Rollup-as-a-service providers such as Caldera, AltLayer, and Conduit. However, with the emergence of L2 and L3 ecosystems, they will default to becoming fragmented. It is already difficult for users to use new platforms - if interoperability, liquidity, and network effects are limited, the situation will only get worse. A unified DA layer as the foundation of each network will make fund flows simpler and attract a wider range of users.

Caldera and other RaaS providers will enable projects to choose the DA layer when building custom aggregations

Avail, EigenDA, and Celestia are the main characters in the DA ecosystem - each serving the same space but taking slightly different approaches in terms of infrastructure stack, execution, and listing.

In terms of technical architecture, Avail, Ethereum, and EigenDA use KZG commitments, while Celestia uses fraud proofs to confirm the correctness of block encoding. Generating KZG proofs, although a very strict method for data availability, brings more computational overhead to block producers, especially as block size increases. On the other hand, Celestia assumes that data can be implicitly obtained through its anti-fraud scheme. As an unfinished computational "work" exchange, the system must wait for a period of time to enter the fraud proof dispute period before nodes can confirm that the block has been accurately encoded. KZG proofs and fraud proofs are both undergoing rapid technological progress; their trade-offs may become more complex, and it is currently unclear which mechanism will strictly outperform the other.

For Avail, they have adopted the KZG commitment scheme, making it very suitable for zk structures. If zk dominates in the future and Celestia relies on optimistic fraud proofs, this may pose a challenge for Celestia. In addition, even if all full nodes are down, Avail's P2P light client network can still support the network; whereas in Celestia's architecture, light clients cannot run without full nodes. Both Avail and Celestia use erasure coding under DAS (distributed storage), dividing data into fragments, increasing redundancy, and allowing data to be reconstructed for verification.

Compared to the technical stacks of Celestia and Avail, EigenDA fully utilizes the existing infrastructure of Ethereum. If data needs to be sent to a Rollup contract to prove its availability, EigenDA will inherit the same finality time as Ethereum. If Rollup fully adopts EigenLayer, finality can be achieved more quickly.

In order to reach consensus, Avail adopts BABE + GRANDPA inherited from Polkadot's SDK, and also uses NPoS (Nomination Proof of Stake). NPoS is used to nominate a group of validators that delegators hope to see elected, while BABE determines who will propose the next block and GRANDPA serves as the finality algorithm for blocks.

Celestia uses Tendermint as its consensus mechanism, allowing users to stake their TIA to receive validator staking rewards. Although Celestia is able to achieve fast determinism through Tendermint, due to its optimistic architecture, there is a waiting period for actual data availability guarantee (users must have time to submit fraud proofs).

EigenDA itself does not have consensus, but there are two mechanisms to ensure the validity of data availability.

Proof of custody: This is essentially an economic security mechanism that ensures nodes store data, but in reality, it cannot guarantee that the data is provided to everyone in the network. If nodes do not comply, they will be slashed, for example, if they cannot prove they own the data.

Sufficient decentralization: Ensuring that the operator set remains decentralized and resistant to collusion is crucial for the normal operation of the network. With a large and independent set of validators, data provision becomes a competition, and many market participants are willing to join. At this scale, collusion is extremely difficult.

One interesting point worth mentioning is that Celestia's active validator set is composed of the top 100 validators with the highest staked tokens, and this threshold may be lowered in the future. Additionally, each of their validators stores the entire dataset. EigenDA, on the other hand, will optimize for nodes (potentially millions in the future) that store only a small portion of the data, so if enough nodes are honest, the data can be reconstructed. The full origin of EigenDA (and more details) can be found in Sreeram's recent post.

Finally, Avail has conducted a useful comparison of the core components of the main DA layer.

There has been a new discussion about the trade-offs for each design. David Hoffman pointed out that Celestia is itself a complete blockchain - a complex stack that requires more than just pure DA. On the other hand, EigenDA is just a set of smart contracts, but it relies on Ethereum, while Celestia and Avail do not.

Celestia team believes that tokens are necessary for security, and EigenDA will eventually need tokens because it is impossible to reduce the availability of on-chain data. They believe that in order to ensure that nodes are honest, data is available, and malicious nodes are punished, the network must be able to verify through incentive structures including native tokens. Here, Nick White of Celestia criticizes EigenDA: "Unless the source chain is forked, validators who retain data will not be slashed - this is highly unlikely because it is Ethereum."

From a branding perspective, EigenDA is a product that is extremely consistent with Ethereum. The EigenLayer team is building it based on EIP-4844 and danksharding - as Sreeram puts it, EigenDA is built as the "only ETH-centric data availability layer". He explains that, by definition, a data availability layer is a modular product, but other DA "layers" are actually the blockchain itself.

It is indeed beneficial to package the DA layer into the blockchain, which provides security guarantees for Rollups running locally on it. However, Sreeram mentioned that the goal of his team in building EigenDA is to create a product that provides data availability services for the Ethereum ecosystem from first principles - a true "Layer" adjacent to the Ethereum ecosystem. He pointed out that separate consensus is not needed here, as Rollups based on Ethereum already rely on the network for ordering and consensus. (Sreeram explained this in a recent Bankless program.)

Avail is built using Proof of Validity and DAS, which enables high flexibility and interoperability in the ecosystem. Their architecture lays the foundation for a scalable framework designed to support services across many different platforms. This unbiased approach allows for greater interoperability and capital flow, and also attracts non-Ethereum-centric ecosystems. The ultimate goal here is to obtain ordered transaction data from all chains and aggregate them into Avail, making it the coordinating center for all web3. To kickstart the network, Avail recently launched node conflict activities alongside its incentive testnet, allowing users to run validators and light clients and participate in network challenges.

The Celestia ecosystem consists of RaaS providers, shared sorters, cross-chain infrastructure, and other components, covering ecosystems such as Ethereum, Ethereum rollups, Cosmos, and Osmosis.

Each of these design choices, whether technical or marketing, comes with interesting trade-offs. Personally, I am not sure if data availability will be a winner-takes-all or commoditized market - instead, there may be an oligopoly market where projects choose the DA layer that best suits their needs. Depending on the type of protocol, teams can optimize for interoperability, security, or preferences for a particular ecosystem or community. If custom use cases aggregate as expected, they will not hesitate to integrate the DA layer - and there will be more than one powerful option to choose from.

This technology - as well as the overall modular narrative - is still relatively new, with Celestia having just gone live and Avail and EigenDA set to enter the mainnet in the coming months. However, the technological advancements in modularism have been remarkable so far (many of which were just ideas a few years ago!). By fundamentally improving the way we build and use blockchain, the DA layer will undoubtedly become one of the core technologies of this and future cycles.