Gate.ioBlogEndgame: Uncovering the Optimal Solution of Blockchain Scaling with Vitalik Buterin
Endgame: Uncovering the Optimal Solution of Blockchain Scaling with Vitalik Buterin
15 December 09:51
On December 6, Vitalik Buterin published an article "Endgame" on his personal blog, planning a reasonable path for the scaling of Ethereum and other major blockchains.
"Large blockchains" refer to those blockchains with large chunks and high trading frequency (up to thousands of trades per second).
Large blockchains can be centralized on block production and decentralized on chunk validation. Therefore, the second layer stake node can be introduced as the validation node in addition to the first layer stake node, that is, the block production node.
After realizing centralized block production + decentralized validation, Fraud Proofs or ZK-SNARKs need to be introduced to enable the second layer staker to directly validate the reliability of the chunk.
Whether it's Ethereum with a Sharding + Rollup structure or a traditional "big block" blockchain, the mode of centralized block production + decentralized validation is always their final destination.
On December 6, Vitalik Buterin published an article "Endgame" on his personal blog, planning a reasonable path for the scaling of Ethereum and other major blockchains to help these blockchains maintain their ability to de-trust and resist censorship when they obtain high scalability, and achieve a better balance in the Impossible Triangle.
In fact, on December 2, a few days ago, on the first anniversary of the launch of Ethereum beacon chain, Vitalik released the subsequent development roadmap of Ethereum on Twitter. In the future, the follow-up work of Ethereum will mainly include the following main parts:
The Merge: merge Ethereum network and completely turn to PoS consensus solution.
The Surge: improve the scalability of Ethereum greatly by using sharding technology.
The Verge: introduce the Verkle tree and other mechanisms to realize statelessness.
The Purge: understand early historical data and other technical issues.
Centralized Block Production + Decentralized Validation
In the article, Vitalik demonstrated the trade-off of "large blockchain" in scaling through a thinking experiment. "Large blockchain" refers to those blockchains with large chunks and high trading frequency (up to thousands of trades per second). Due to the large chunk capacity, the size of the entire blockchain ledger will also expand rapidly, so that most nodes cannot meet the hardware requirements for running the blockchain and exit the network. This "large blockchain" is often highly centralized, and there may be only hundreds or even dozens of nodes with block production capability and nodes running the entire java_script_. So is it possible to make this blockchain that does not perform well in decentralization also have a high capability to de-trust and resist censorship?
Firstly, Vitalik believes that these blockchains can be centralized on the block production and decentralized on the block validation. Therefore, the second layer stake node can be introduced as the validation node in addition to the first layer stake node, that is, the block production node. This second layer stake has fewer resource requirements and can realize the distributed validation of chunks at a low cost. Specifically, the trades in each chunk need to be divided into 100 groups, and each group is used as the root of a Merkle Tree or Verkle Tree. In this way, the validator only needs to view the top node of Merkle Tree to complete the validation of a large number of trades at one time. Even if there is a problem, the validator can search down and quickly find the location of the questionable trade. After that, each second layer staker will randomly obtain one of the 100 groups of trades. One chunk will be accepted only if more than 2 / 3 of the second layer staker assigned for validation pass the validation.
Vitalik drew a flow chart describing the process of "central production block" + "distributed validation". He thought this method was "not beautiful" and "aesthetically ugly", but it did provide the system with the ability to resist censorship. Even if a layer 1 staker (block producer) intends to attack or censor the chunk content, the most he can do is to leave the network and temporarily make the Ethereum network unable to produce blocks until the next honest node is selected.
Secondly, after realizing centralized block production + decentralized validation, it is also necessary to introduce Fraud Proofs or ZK-SNARKs, so that the second layer staker can directly validate the reliability of the chunk. The cost of these two technologies is very low. Among them, ZK-SNARKs allows to directly prove the reliability of chunks in cryptography. The principle of Fraud Proofs is that once the trade assigned by the node is found to be wrong, the Fraud Proofs can be broadcast.
Thirdly, introduce DAS Technology (data availability sampling). DAS is closely related to Sharding and Rollup technology. We know that Sharding technology not only eliminates the need for nodes to store all data, but also provides Rollup with data availability. Using DAS technology, users running lightweight nodes can also validate the openness and transparency of blockchain content without downloading all data.
Fourthly, add secondary trading channels to resist censorship. The nature of blockchain resisting censorship can be described as: "no one can prevent a trade on the blockchain or add code to the smart contract. Users can perform whatever they want on the blockchain". Vitalik also provides a possible secondary trading channel scheme, which allows the second layer staker (validator) to submit a trading list and must include the next chunk.
Sharding and Rollup
Ethereum 2.0 itself cannot solve all the problems of scalability, but it provides a good platform for others to propose and solve problems, such as Arbitrum (Layer 2 solution), Optimism and Zksync. On Ethereum, DAS technology makes Sharding possible, and Sharding provides Rollup with data availability. The data after Rollup will be distributed to each major Sharding. Rollup nodes or block production nodes require high processing capacity, and Rollup mechanism also ensures the effectiveness of chunks. If a Rollup is successful and occupies the whole Layer 2 tracker, it also constitutes a capability of centralized block production + distributed validation.
If no single Rollup technology can occupy the whole market, all kinds of Rollups will coexist. Then, due to the existence of cross-domain miner extractable value opportunity (MEV), multiple Rollups have complex trading and arbitrage opportunities, which will eventually lead to the situation that a small number of people seek MEV to control the whole blockchain system.
Finally, we can conclude that Vitalik thinks whether it's Ethereum with a Sharding + Rollup structure or a traditional "big block" blockchain, the mode of centralized block production + decentralized validation is always their final destination, which is what he calls the "Endgame". However, in view of the immature technologies such as Shardind and Rollup, it may be several years before this "Endgame" comes.
Author: Edward.H, Gate.io Researcher
*This article represents only the views of the researcher and does not constitute any investment advice.
*Gate.io reserves all rights to this article. Reposting of the article will be permitted provided Gate.io is referenced. In all other cases, legal action will be taken due to copyright infringement.