Author: Hao Tian; Source: Chain View
Many people think that ZK-Rollup is the endgame of Ethereum layer2, but as I delve deeper into the research, I find that it is not the ZK-Rollup form that can be called the endgame, but the ZK technology itself. Because in theory, OP-Rollup can also use ZK technology to reduce the time limit of the 7-day challenge period, and even eliminate the MPC multi-signature governance defects. How to understand it? This article will take @MetisL2 and the technical hardcore project @ProjectZKM behind it as an example to discuss:
In my previous article, I also explained that the "ZK technology" for layer2 expansion has only been tapped to a limited extent, and there is still a lot of room for full-chain trustless interoperability. In terms of layer2 application scenarios, it can directly bring two major impacts: 1) The cross-chain funds of layer2 and layer1 will be able to directly make deterministic delivery of secure assets based on ZK underlying technology. Using ZK technology, OP-Rollup can significantly shorten the 7-day waiting challenge period required for fund withdrawal; 2) The Prover verification system of layer2 and the Rollup verification contract of layer1 can realize the construction of a trusted environment across the main chain. In theory, there is no need to use the much-criticized "rule of man" model such as centralized or semi-centralized MPC multi-signature to ensure security; In fact, @MetisL2, as a layer2 chain based on the OP-Rollup technical framework, has already realized the above two dimensions of layer2 underlying architecture paradigm transition through ZK underlying technology: On the one hand, the hybrid Rollup framework of OP-Rollup+ZK-Rollup optimizes the user experience of waiting 7 days for withdrawing coins, which can release liquidity faster; On the other hand, Metis layer2 is used to experiment with the goal of gradually implementing @ProjectZKM's "trustless" framework for cross-chain interoperability, using ZK as the decentralized coprocessor built on the infrastructure as a trusted subject to generate proofs, verify proofs, and manage the communication and trust mechanism between cross-chain blockchain nodes. How to do it specifically? - Hybrid Rollup: Normally, a layer2 is either an OP-Rollup Rollup structure based on the optimistic 7-day challenge period proof, or a ZK-Rollup single-block Finality proof structure based on ZK-SNARKs.
The former has a fast Batch transaction rate and low handling fee, but the disadvantage is that when users withdraw funds from Layer 2 to Layer 1, they need to wait for a 7-day challenge waiting period, because according to the optimistic Batch essence of OP-Rollup, only if the withdrawal funds are not challenged and traced back within 7 days, the withdrawal of this fund can be considered safe in theory;
The latter ZK-SNARKs' Proofs transmission structure can realize Finality asset security confirmation in a single block, but the disadvantage is that in addition to the DA cost of Batch transactions, the Prover system calculation and verification cost will be added, and the 2-layer ecological construction has a high starting threshold for the developer team, and the ecological development speed is slow.
So, how can we apply the ease of use of the OP-Rollup overall architecture while being free from the limitation of its 7-day challenge period (locking liquidity)? The answer is: the transactions from the 2nd layer batch back to the main network are processed in different "channels", with Common transactions going through the OP-Rolluo pipeline and special Withdraw transactions going through the ZK-Rollup pipeline. In this way, the strengths of both OP and ZK are adopted to form a compromise optimal solution.
Metis has implemented the hybrid Rollup solution into the normal operation of its chain, and the "new" hard-core ZK technology underlying project @ProjectZKM incubated and created by the Metis technical team provides it with this hybrid Rollup capability. The Metis team discovered some inherent normal problems in the operation of layer2. In addition to taking the lead in overcoming the decentralized sequencer problem, it also aims to solve a series of problems such as "MPC semi-centralization" and "locking liquidity" in the development of layer2 through ZKM;
At first glance, you may think it will be very abstract. The reason is explained in my previous article about ZK technology. This type of service is currently at the upstream of the layer2 technology service supply chain. Although hard-core, it has not been widely adopted, so it sounds very unfamiliar. Like RISC Zero, ZKM aims to provide more upstream security consolidation services for layer2 projects based on the ZK underlying technology framework.
Currently, ZKM provides the ZK-Rollup pipeline service for the hybrid Rollup required by layer2, and there is also an Entangled Rollup protocol at the bottom to provide unified liquidity management services.
In fact, upstream ZK technology service providers such as @ProjectZKM, @RiscZero, and @SuccinctLabs are promoting the popularization of similar technical solutions, thereby helping OP-Rollup and ZK-Rollup to further reduce the differences, while reducing the loss of liquidity caused by the 7-day challenge waiting when there is no Battle-Tested challenge at all. According to incomplete statistics, Metis, Fraxchain, Aztec, Ola, etc. have been exploring and practicing the implementation of such hybrid Rollup solutions.
-Based on ZK-free cross-chain bridge trust mechanism:
Most layer2s are now criticized for the problem that the Update proxy contract in the mainnet Rollup contract is controlled by the semi-centralized MPC security governance committee, which causes most layer2 security to remain at the Stage 0 stage mentioned by Vitalik.
Upstream ZK technology service providers such as ZKM and RISC Zero can theoretically enable each chain with smart contract functions to achieve a decentralized trusted environment for MPC-free multi-signature cross-chain bridges.
The principle is very simple. ZK-SNARKs allows the nodes that maintain consensus on the two chains to establish direct communication. After receiving the transaction and ZKP proof submitted by the B-chain node, the node of chain A can independently verify the validity of the ZKP proof, and then accept the validity of the Proofs submitted by chain B. The entire process is completely implemented by the zero-knowledge proof algorithm, and does not require a third-party human subject to control.
As for the coprocessor that schedules the communication between nodes, it can be implemented with a decentralized open source architecture, thus completely solving the "centralization" problem of Rollup cross-chain.
Since Ethereum has the completeness of smart contracts, in theory, Ethereum's layer2 has the basic conditions for cross-chain based on ZK technology. Ideally, layer2 uses the same ZK technology framework to deploy cross-chain solutions, and other layer1 chains and even layer3 chains are also incorporated into this network communication architecture. Wouldn't the Ethereum public chain in the future become a universal settlement layer in a full-chain environment?
The reason why most layer2 project parties are unwilling to give up MPC multi-signature subject management is mainly due to security control. In my opinion, the fundamental reason is that the popularization of basic component services such as decentralized sequencer, decentralized prover, and decentralized zkBridge at the entire layer2 level is not in place. In this case, retaining a centralized or semi-centralized security governance committee has become a safer governance method.
However, in the long run, as layer2 becomes popular in decentralized components, layer2's security governance method will certainly be magnified and criticized, and a cross-chain settlement solution based on the ZK underlying layer will become a necessity. The subsequent development potential of the ZK technical infrastructure track can be imagined.
The above
It is no exaggeration to say that Ethereum layer2 is currently in a development dilemma: either rely on the application ecology of layer2 or layer3 to produce a phenomenal king bomb project to boost everyone's market confidence in layer2, or it can only rely on further improving the decentralization of key components such as Sequencer, Prover and zkBridge to continue to enrich the technical narrative. The latter will inevitably need a set of underlying frameworks based on ZK technology to connect, without the distinction between OP-Rollup and ZK-Rollup, when the Endgame feature of ZK will truly appear.
Only by following this path, the development direction of layer2 will not be falsified, and based on this, it can also extend the development of ZK hardware device acceleration; ZK+DePIN PC, mobile terminal, IOT and other hardware device support; ZK anti-tampering cloud computing data protection; ZK decentralized Prover system construction and computing power resource optimization and many other narrative directions.
In fact, there are already many similar projects at the intersection of AI, ZK, and DePIN tracks that are gradually emerging.
The same is true for Metis's development ideas in layer2 and the original intention of the hardcore ZKM project, making ZK technology more universally applied to every link of the global layer2 market construction; in addition, my personal investment research has also noticed some projects, such as @cysic_xyz, which continues to work on ZK hardware and chip acceleration, and @ola_zkzkvm, which is gradually landing and practicing on the ZK Prover decentralized computing power system.
In short, I want to reiterate my point of view that the current application of ZK technology in the layer2 field is just the tip of the iceberg. When ZK technology is further popularized in the fields of hybrid Rollup, ZK cross-chain bridge, ZK hardware acceleration, etc., I believe it will also bring new vitality to the layer2 market.
After all, it will take a long time for the entire layer2 market to accept this ZK bottoming upgrade paradigm.