Over the past few years of blockchain development, the layer paradigm has been changing. Recently, V God emphasized the significance of Layer3, because the architecture can optimize and improve the network , provided that they have different functions. But with the plethora of multi-chain solutions available, can L3 compete with current L2 and L1? The existing L2 has not solved the blockchain trilemma (one of the biggest narratives in Web3). Next, the Odaily Planet Daily will take everyone to explore whether L3 can really solve this nearly ten-year scalability problem.
Introduction to Crypto Layers
L1 refers to a blockchain such as Ethereum or any blockchain that can stand on its own. In other words, those chains that are not built on top of another chain and can function completely independently. L2 is a third-party integration that works with the L1 blockchain. Essentially, they are extending the underlying L1-based solution, adding security and scalability.
So what is L3? Are we just building L3 on top of L2 for performance and security? Simply put, it is to change the scaling solution to get a higher level scaling solution. Ethereum is so congested that people want to add another layer to prevent congestion. But each layer has some new features that don't allow stacking, which means improving scalability only once.
Recently, Starkware came up with a more viable idea for L3, adding layers for different purposes. As Vitalik highlighted, this approach might actually work, but let's first look at why we can keep scaling by stacking rollups on top of rollups.
Problems with Rollups
rollup executes transactions outside of the main Ethereum blockchain and sends data back to the main Ethereum network. Rollup can also be considered as a scaling technology that mainly solves two bottlenecks—computing and data. Computations refer to Fraud Proofs, or SNARKs, which rely on a small number of participants to process and validate a block, requiring others to perform only a fraction of the computation to ensure validity.
This means that SNARKs can be extended almost unlimitedly, for example, a SNARK embedded in a SNARK, and a SNARK is a subset of SNARKs in the SNARK ecosystem.
However, the data is a bit more complicated. The purpose of rollup is to compress on-chain data, which is why there is a lot of talk about a zK proof solution, as it compresses almost 8x more data. The problem is that rollup still requires data available on-chain to be open-sourced and for users to access and verify the data. This allows them to independently compute the state of a rollup and join as attestors when the previous rollup was inactive.
Unlike computation, data cannot be compressed repeatedly, only once. So you can embed rollups as a subset of rollups in the rollup ecosystem, but that doesn't improve scalability.
Starkware's Vision for L3
The current version of Starkware uses layers for different purposes. Essentially, if a rollup compresses the data by a factor of 8, a rollup on top of a rollup compresses the data by a factor of 64.
V God took Starkware's L3 as an example to propose 3 "L3 visions", which are mainly composed of 3 main purposes, emphasizing again that L2 and L3 need to have different purposes. Highlights include:
-
L2 can be used for extensions, while L3 is used for custom features such as privacy;
-
L2 is for common extensions, while L3 is for custom extensions;
-
L2 is used for trust-free scaling, also known as rollup, while L3 is used for less-trusted scaling solutions.
-
Is L3 really more efficient than L2?
I think yes, it's cheaper than the existing L2 model. Deposits and withdrawals can be made between L2 and L3 without interacting with the underlying chain (L1). This means tokens can be transferred between L2 without going back to the main chain or L1. The problem is that deposits need to wait for the fraud prevention window. For example, if a token originates from Ethereum, withdrawals to Ethereum from L2 with the base or wrapper contracts would require huge delays, whereas deposits are instant.
This is where zK-rollups come into play, since they are much faster and don't require such a long wait time for security reasons. In addition, the cost of rollup is relatively low. However, submitting batches of transactions to the chain has a high fixed cost. Rollup can wait until there is a large demand for transactions before submitting a batch, which forces users to wait for high-security confirmations. This approach is not feasible for corporate and retail businesses.
Imagine waiting 12 hours for an instant transaction, which is not feasible for daily transactions. This means there are two options, wait longer for low cost, or wait shorter for high cost. L3 can solve this problem, because zK-rollup in zK-rollup will greatly reduce the cost by nearly 20 times, which means that transactions can be sent quickly while still maintaining a high degree of security.
However, if the demand is still low, it may cause problems, but we can optimize it by adding another layer (L4) on top of L3. If we use L3 for custom transactions, we can create L4 as a one-off layer only used when traffic subsides.
Summarize
In summary, an L3 scaling solution that stacks the same layer on top of itself won't work. L2 leverages the security of the underlying L1 while improving performance; L3 can extend the capabilities of the current blockchain by implementing "custom extensions"; combined with new programming languages, L3 may be the real factor driving mass adoption. While rollups may not be the best solution, leveraging their technology can help improve the overall functionality of any blockchain network.