Blockchain circle

One stop hot information platform

About us:

Blockchain circle provides the latest information about blockchain, digital currency, digital wallet, exchange, metauniverse, bitcoin, Ethereum, contract, financial management and so on, and always pays attention to the latest market...

Gu Lumi observed: why is "decomposition" the first principle of blockchain optimization?

Time : 12/05/2022 Author : 71pda3 Click : + -
        In the past two weeks, I have spent a lot of time to understand the latest chain level innovations, and in my opinion, "disaggregation" is the first principle guiding these innovations. This word comes from & lsquo; Modular blockchain & rsquo; (my latest article explains why it is decomposition rather than modular blockchain). There are still some differences between decomposition and Modularization:. The logic behind it is very easy to understand. By enabling each unit to perform the simplest operation, the efficiency of each unit can be maximized. In the AMA studied by Ethereum foundation No. 8, Justin said & lsquo; Ethereum is becoming more and more modular:.
        The strategic approach to reducing complexity is modularization. (see this article on encapsulation and systemic complexity). The good news is that Ethereum is becoming more and more modular.: The consensus layer is largely encapsulated. Peter only means that the execution work (formerly within the purview of the execution team) has been outsourced to the broader rollup community.: The separation of data (danksharding) and execution (rollups) means that the execution (formerly within the purview of the execution team) is outsourced to the broader rollup community: The complex low-level bls12-381 cryptography is encapsulated in the library. For example, Peter can interact with the blst application program interface when processing the verkle tree.
        : proposer builder separation (PBS) allows non consensus critical builder logic to be separated from consensus critical proposer logic. I would like to see the emergence of two types of execution clients: the proposer client for verifiers and the builder client for the MeV (miners can extract value) industry: In the context of enshrinedzkevm (zero knowledge proof Ethereum virtual machine), the checker logic of non consensus critical can be separated from the snark verification logic of consensus critical.
        Similarly, I hope that the client can be further standardized and modularized in enshrinedzkevm. But you should notice that Justin's & lsquo; Modularization & rsquo; It has gone far beyond what most people are discussing & lsquo; Modular blockchain & rsquo; Category of:. Parallel computing is an extension technology that has been proven by web2, but there are also some preconditions for the implementation of this technology, and the parallel technology naturally does not match the current mainstream single block chain. Transactions processed in parallel cannot have any correlation, but there are too many correlations between transactions in a block.
        Altlayer is a highly scalable application specific and pluggable execution layer system, which can obtain sufficient security from the underlying L1 / L2 (layer 2 network). If you ask me what is the difference between altlayer and other mainstream layer 2 networks, I will say: altlayer is an automatic expansion solution, while other solutions are more "immutable". The correct general framework of the blockchain is becoming clearer and clearer, but we still need to carry out more optimization to make the blockchain better. How should we optimize the existing blockchain?.
        Recently, there have been a lot of discussions about who is expected to become the biggest competitor in the next bull market of Ethereum. Is Celestia, Aptos and Sui the next killer of Ethereum?. Four layers: the decomposition of single blockchain into four layers has proved its value. (I understand that we can divide the single blockchain into more layers, but to make it easier to understand, I use the term "four layers" here). Celestia's design and DAS (data availability sampling) are the best choice for the data availability layer. (theoretically, it is safer than the other options mentioned above). Parallel computing is also valuable for further optimizing the transaction consensus layer, and is theoretically feasible.
        Celestia is the only choice for transaction order consensus (currently, the design of simultaneous processing of data availability layer and transaction number consensus by a group of people is the most effective). Aptos and Sui are designed to have the best performance in global state consensus, with parallel computing and unique P2P messaging. Vitalik jokingly said & lsquo; We should cancel Segmentation & rsquo; Go to @ polynya's latest & lsquo; 4844andDone’, There is a lot of information about & lsquo; Reduce the complexity of blockchain & rsquo; Discussion.
        My view is more like & lsquo; It is still not the time to simplify: If we review the development history of the cloud computing industry, in the early 5-10 years, cloud computing services have become more and more useful, but more complex. In recent years, we have seen more simplification of cloud computing services, such as no service, IAC, etc. I still believe that we can see more decomposition innovations in this bear market. When I encounter these innovations, I will share my ideas with you in time.
Previous:Development of Beijing blockchain system
Next:No more

Related articles:

© 2005-2032 | Blockchain Circle & & All Rights Reserved    Sitemap1 Sitemap2 If there is infringement, please contact us at: