Vitalik Buterin: Ethereum roadmap centered on rollup

Editor's Note: This article comes fromBabbitt Information (ID: bitcoin8btc), Author: Vitalik Buterin, Compiler: Free and Easy, released with authorization.

Editor's Note: This article comes from

Babbitt Information (ID: bitcoin8btc)

Babbitt Information (ID: bitcoin8btc)

Written in the front: Facing the growing demand for expansion, Vitalik Buterin, the co-founder of Ethereum, recently proposed a new roadmap for Ethereum centered on rollup. ENS domain names, applications, etc. are also entirely on layer 1, and all these things will have to change, and we need to adapt to a world where users' main accounts, balances, assets, etc. are completely migrated to layer 2 networks such as rollup.

Last week, the Optimism team announced the first phase of its testnet, as well as a roadmap for the mainnet. They are not the only ones, Fuel is moving towards a testnet launch and Arbitrum already has a testnet. In the field of ZK rollup, Loopring, Zksync and Deversifi based on Starkware technology have been launched and have users on the main network. With the OMG Network's mainnet beta running, plasma is also moving forward. Meanwhile, Ethereum 1.0 gas prices are climbing to new highs, to the extent that some non-financial dapps are being shut down, while others are opting to run on testnets.

The Ethereum 2.0 roadmap provides scalability, and the early phases of Ethereum 2.0 are fast approaching, but the base layer scalability of the application will not be achieved until the last major phase of Ethereum 2.0, which is still a few years away, Even more ironically, the availability of Ethereum 2.0 as a rollup data availability layer appears in phase 1 (phase 1), which will land well before phase 2. Taken together, these facts lead to a particular conclusion: the Ethereum ecosystem will likely migrate to rollup (plus some plasma and channels) as a short- and medium-term scaling strategy.

If we start from this premise, we can see that it leads to some specific conclusions about the prioritization of Ethereum core development and ecosystem development, which in some cases differ from the current path. So what are these conclusions?

• EIP 2929secondary title

• EIP 1559Short term: Advancing Ethereum 1.0 for Rollup

• In the short term, the main result of this work is to focus the expansion of the Ethereum base layer on expanding the capacity of data blocks, rather than the efficiency of on-chain calculations or IO operations. The only determinant of Rollup scalability is how much data the chain can hold, and any increase beyond the current ~60kb/s will help further improve Rollup scalability.

• There are still some issues at the base layer:

, to ensure that the blockchain is resistant to DoS attacks at the current gas level;

New elliptic curve precompilation that fully supports what people want to achieve with ZK rollup;

Hex tree -> binary tree changes and other changes to improve support for stateless clients (since stateless clients are valuable regardless of how the blockchain is used);

The importance of account abstraction is not very high, because it can be implemented on L2 without caring whether L1 supports it or not. And other "smart base-layer features" have been lowered in priority.

• secondary titlehere；

• Short term: Adjusting infrastructure for Rollup

• Currently, user accounts are on L1, ENS domain names are on L1, applications are entirely on L1, and so on, and all those things will have to change. We need to adapt to a world where users' primary accounts, balances, assets, etc. are entirely in L2. Here are some things to be aware of:

• here

The Layer 2 protocol should be built into the wallet, not like a web dapp. Currently, L2 integrations into dapps/quasidapps (such as Gitcoin's zksync integration) require users to fully trust the dapp, which greatly reduces security. Ideally, we'd like L2 to be part of the wallet itself (metamask, status, etc.) so we can keep the current trust model. This support should be standardized, so that applications that support zksync payment will immediately support zksync built-in Metamask, zksync built-in Status, etc.

We need to do more work on cross-L2 transfers to make the experience of transferring assets between different L2s as close to instant and seamless as possible;

Standardize more explicitly on Yul or a similar intermediate compiled language.

Advantages of Economic Sustainability Centered on Rollup

It is an unavoidable question that a crypto project must be financially sustainable, and in 2020, that means millions or even tens of millions of dollars. Some of this could be covered by public goods funding entities such as Gitcoon Grants or the Ethereum Foundation, but these mechanisms are not large enough to cover this level of funding. However, layer 2 projects can cover it by launching their own tokens—of course, the premise is that the tokens are backed by real economic value (that is, L2 must be able to capture fees).

Another benefit of a rollup-centric roadmap is that it leaves room for L2 protocols capable of collecting fees/MEV that can directly or indirectly fund development (by supporting tokens that fund development). The Ethereum base layer has another important requirement, credible neutrality, which makes public goods financing within the protocol difficult, but L2 has its own public goods financing mechanism (or Gitcoin grants), which is more controversial than much smaller. Therefore, opening up this space is a good strategic move for the overall long-term economic sustainability of Ethereum.

Funding issues aside, the most creative researchers and developers often want to be in a position of influence on their own island and then argue with others about the future of the Ethereum protocol. In addition, there are many existing projects trying to create various platforms. A rollup-centric roadmap provides a clear opportunity for all of these projects to become part of the Ethereum ecosystem while still maintaining a high degree of native economic and technical autonomy.

• secondary title

• long term focus

• Beyond these short-term concerns, a rollup-centric roadmap could mean a reimagining of the long-term future of Ethereum 2.0: as a single high-security executive shard that everyone can process, plus a reliable Extended data availability layer.

• To understand why this is the case, consider the following:

Today, the TPS of the Ethereum mainnet is about 15;

If everyone turns to rollup, we quickly have about 3000 tps;Once the Ethereum 2.0 phase 1 appears, and the rollup is transferred to the Ethereum 2.0 shard chain for data storage, we will theoretically reach a maximum of about 10,000 tps;Eventually, Phase 2 will appear, which will bring Ethereum 2.0 shard chains and native computing, which will bring us about 1000-5000 tps;

In my opinion, when Phase 2 finally arrives, basically no one will care about it. Whether we like it or not, everyone has adapted to a rollup-centric world, and by then it will be easier to continue down that path than trying to get everyone back on the base chain (going back and No benefit, and 20-100x lower scalability).

This is equivalent to an Ethereum 2.0 "" approach, where the base layer is scaled down and focused on doing a few things well, namely consensus and data availability.。 

In fact, this might be a better position for Ethereum 2.0, since shard data availability is much safer than shard EVM computation. While dishonest majority proofs for sharded EVM computations require fraud proofs, which require a strict and potentially risky two-epoch synchrony assumption, data availability sampling (if using zero-knowledge proofs ZKP or polynomial commitments) is safe in the asynchronous case of.

• This will help distinguish the Ethereum 2.0 main chain from other sharded L2 chains, where Ethereum 2.0 will be a base layer that is powerful enough to enable

• functional escape velocity

• secondary title

• Stagger block times across different shards, so at any given time, some shards are scheduled to propose a block within a few hundred milliseconds. This enables rollups operating across multiple shards with ultra-low latency without the risk of the chain itself having ultra-low latency.

Improve and solidify its consensus algorithm;

Tweaking the EVM to be more friendly to fraud-proof proof verification (for example, this could imply some sort of "framework" functionality that prevents code from spilling out of the sandbox), or allowing SLOAD/SSTORE to be remapped to use something other than account storage other content as its data source;

ZK-SNARKing anything;