Why are there so many Layer2 solutions, but no one uses them?

Editor's Note: This article comes fromBlock beats BlockBeats (ID: BlockBeats)Editor's Note: This article comes from

Block beats BlockBeats (ID: BlockBeats)

Block beats BlockBeats (ID: BlockBeats)

, Author: 0x88, reprinted by Odaily with authorization.

On-chain scaling is not a “new” problem.

As early as 2017, Ethereum and Bitcoin were congested and unable to meet the needs of users. Since then, development teams have successively proposed corresponding solutions, but the ensuing bear market has reduced people’s enthusiasm for trading. , This type of program also faded out of the investor's field of vision.

Until May of this year, with the popularity of liquidity mining and Uniswap, people discovered again: Gas price soared to 500 Gwei, and the number of unconfirmed transactions reached more than 140,000. Someone ridiculed that the performance of Ethereum is not enough to provide a normal use of a DeFi project. Therefore, the issue of on-chain expansion has become a hot topic again.

When it comes to expansion, we have to mention the impossible triangle problem of the blockchain. For a public chain, security, decentralization, and scalability cannot be achieved at the same time. Theoretically, in order to ensure the degree of decentralization and security, each piece of information on the chain needs to be authenticated by all nodes to reach a consensus, which means that the throughput of a chain is equal to the throughput of a node.

Off-chain scaling is the only relatively mature technology. The main logic of this solution is to minimize the burden on the main chain of the first layer, and complete the application logic on the second layer. The reason why this solution can be realized is that not all transactions need to reach a global consensus, and not every transaction needs to change the global state. A layer of public chain only needs to be responsible for ensuring public verifiability.

It can be seen from the above figure that the Layer 2 projects of Ethereum are already very rich, but Ethereum will still be blocked. Why are Layer 2 applications rarely used?

In order to gain a deeper understanding of the development status and problems of Layer 2 ecology, Rhythm BlockBeats discussed with JD (Jaynti) Kanani, founder and CEO of Matic, the leading Layer 2 project, and Dr. Dong Mo, co-creator of Celer Network. In the following, we try to answer a core question from the most front-line perspective: Why does Layer 2 seem to have achieved remarkable results, but why is it still so far away from us?

secondary title

Ecological development of Layer 2

Even if you have never experienced the convenience brought by Layer 2, everyone has more or less heard of Layer 2 projects or tokens. In terms of technical architecture, there are a lot of Layer 2 solutions, such as zkRollup suitable for large-scale transfers, Optimistic Rollup being done by Optimsm and Offchain labs, Hybrid Rollup and state channel network proposed by Celer, and some projects still insisting on Plasma.

After the rise of the DeFi boom, Dr. Dong Mo of Celer Network clearly felt the market's change in the underlying infrastructure: "Layer 2 ecology has been steadily advancing, and developers are becoming more and more active. With the emergence of Ethereum's congestion problem , Developers’ attention to Layer 2 is also increasing.”

Not only are developers concerned, but more importantly, different public chains are also laying out their own Layer 2 expansion ecology, including Polkadot, Oasis Labs, Nervos, etc. So, among so many proposals, what is the biggest technical problem faced by these projects?

In the middle of different paths, there are various roadblocks. Matic's JD opines: "The mass exit problem is intractable for Matic's Plasma solution. The PoS data availability layer in Matic's Plasma + PoS hybrid architecture minimizes the exit problem. Applying multiple validators instead of a single node Authentication reduces fraud and data availability issues to near zero."

Dr. Dong Mo told Rhythm BlockBeats: "Although Celer's Hybrid Rollup does not have the problem of "who decides the order of transaction packaging and connection with the main chain" in Opitmistic Rollup, and "how to design the incentive mechanism of block nodes to avoid transaction censorship" problem, and the problem of "how to speed up the delay of transaction confirmation as much as possible", but the problem of "how to help light nodes monitor the validity of transactions" still exists."

But many people may be wondering: Why is the Ethereum network still so congested under the impact of the DeFi boom even though there are already so many Layer 2 solutions? JD believes that there are two main reasons for this split. One is the developer problem. "For most developers, they want to migrate their applications to Layer 2 without changing the original code. Because the code Making any changes will involve considerable audit and maintenance costs. But if the Layer 2 project party can make it easy for application developers to complete the migration, this will promote developers’ willingness to migrate to a certain extent.”

The second is a user-side problem. "For users, it is a very difficult issue to choose whether to use the Ethereum main chain or Layer 2 applications. Users need to understand the pros and cons of using both, or in other words, the Layer 2 project can directly allow users to realize the non-intuitive interaction with Layer 2. 2. Without reasonable incentives, transferring assets between layer 1 and layer 2 is extremely challenging for users.”

In addition, Dr. Dong Mo also believes: "The process of entering and exiting Layer 2 is often complicated. This is a threshold for technology and users, and it is also a threshold for ecology. Layer 1 infrastructure (Infura, browser, light client, transaction Support, etc.) are relatively complete, and in the process of migrating applications to Layer 2, there is still a need for corresponding supporting infrastructure, these things are not difficult to do, but it will take a little time.”

secondary title

Are DeFi projects suitable for deployment on Layer 2?

Let's go back to the question of DeFi.

The multi-frequency and complex interactions in DeFi applications are the essential reasons for the congestion of Ethereum this time, so are DeFi applications suitable for deployment on Layer 2? Dr. Dong Mo believes: "Theoretically, it is very suitable. Layer 2 can greatly reduce transaction costs. But the core problem here is still the two points mentioned before: ecological support and the test of time."

In addition to the endogenous ecology, the external ecological environment is also very important. As you can imagine, if there is no support from the trading platform: if you dig a layer 2-based sweet potato, if you have to wait a long time to switch back to Layer 1, and then send it to the trading platform to complete the transaction, the market price may have already changed.

In addition, although there is no security problem in theory, in practice, whether it is engineering or production-level product quality, it takes time to test and the accumulation of confidence.

It is understood that some DeFi projects (EasyFi) have recently migrated to Matic, including the above-mentioned decentralized derivatives trading platform dYdX, which has also migrated to StarkWare, a Layer 2 solution based on zero-knowledge proof. JD believes that in the long run, DeFi projects will eventually migrate to Layer 2. So for each type of project, what kind of Layer 2 solution should DeFi projects choose?

secondary title

How to choose a Layer 2 solution for DeFi?

For this question, we can first make a simple comparison of the advantages and disadvantages of zk Rollup, Optimistic Rollup and Hybrid Rollup.

According to Dr. Dong Mo from Celer Network, the characteristic of zk Rollup is that after the transaction data (calldata) and zero-knowledge proof of the second layer block are uploaded to the chain, the validity of the transaction can be confirmed immediately. Due to the characteristics of zero-knowledge proof, correctness verification is a very simple operation, which can be completed directly on Layer 1, and the confirmation timeliness is relatively high, which is suitable for transfer.

Of course, this solution also has disadvantages. "First of all, there is no general solution for the construction of the verification link, so there is currently no good way to achieve a very generalized virtual machine logic. Simply put, zk Rollup must be customized for each use case; secondly, the verification of program correctness It is relatively complicated, and the polynomial curcuit needs to be verified; finally, the second-layer packaging node has a heavy burden and high cost, and it takes a long time to calculate the zero-knowledge proof, and the user experience of delay is still relatively poor.”

The main disadvantage of Optimistic Rollup is that its own transaction latency is a bit high. This latency is even higher than Layer 1, but it has the advantage of being able to support arbitrary smart contract logic with unlimited participants. A lot can be packaged at one time. In Dr. Dong Mo’s opinion, “there is no good solution to the transaction ordering and packaging mechanism, but a more complicated miners residual value auction (MEV auction) mechanism is needed to select packaging nodes, and censorship is prone to occur. question."

In addition, Celer's Hybrid Rollup solution is a bit like a hybrid car. It combines Optimistic Rollup and Celer's SGN. and payment issues,” said Dr Dong Mo.

Of course, not all interactions have such high security requirements, only when it comes to transactions, the operational security requirements are relatively high, but the data interaction of many applications does not require such high security requirements. Therefore, we can think about it this way: Many operations that do not require high security can be confirmed by side chains. At the same time, multiple Rollups can be combined on the same side chain to achieve the effect of application layer sharding and improve composability. Even if there is a problem in the confirmation, it can still be confirmed through the final main chain for subsequent error correction.

After clarifying the advantages and disadvantages of various Rollup solutions, Rhythm BlockBeats believes that various DeFi applications can choose the expansion solution that best matches the project characteristics according to their needs: