Huang Haiquan, head of Wuzhen·JD Chain: The new generation blockchain is expected to be compatible with public chain and enterprise-oriented application scenarios.

From November 8th to 9th, the “2019 World Blockchain Conference·Wuzhen” hosted by Babbitt was held at the Wuzhen Internet International Convention and Exhibition Center.

This afternoon, at the "New Potential, New Decade" forum, Huang Haiquan, the chief architect of JD Blockchain and the head of JD Chain, gave a speech entitled "Exploration and Prospects of the New Generation Blockchain".

Bl (4)

The following is the full text of the speech, published by Babbitt:

My sharing is divided into three parts:

1. Blockchain soul three questions, but these three issues are critical issues in the blockchain field, coincide with Huawei's Zhang Zong, indicating that our industry has consensus on these three issues;

2. Jingdong Digital Science and Technology in the blockchain technical architecture and practice;

3. JD Chain open source project.

First, the blockchain soul three questions

1. What is the blockchain?

I think there are two perspectives to see "what is the blockchain": one is a static perspective. This picture undoubtedly illustrates what "blockchain is", that is, a combination of some off-the-shelf technologies, so some people call the blockchain a distributed database, and this is not wrong, but it may narrow down the zone. A description of the value of the blockchain.

If you look at this issue from a development perspective, you can say "What is a blockchain?" To answer this question, I think we can look at the nature of the blockchain. We want to explore the essence. I think the core point is that when we are talking about the blockchain, what is our core? Obviously everyone expects us to have a very reliable data protocol that can protect personal information and our value. We can exchange directly with each other, without going through an intermediary, etc. This is obviously our blockchain. The core appeal. The exploration of this value does not have to be constrained by the specific form of the blockchain we are seeing now. So I said "What is a blockchain?" I think that we can satisfy these demands, we can all call it a blockchain.

2. Why do you need a blockchain?

This question is very interesting. We started from the deep investment in 2016 to explore the blockchain field. We have been constantly questioned by customers. In fact, the other meaning of this question is what is the value of the blockchain, so that we should use it. Although we rarely see the existence of a mature business closed-loop case today, it does not prevent us from exploring the value impact that it may have in the future based on today's blockchain profile. This is a summary of some of the value impacts we have made. There are two big aspects. Let's talk about expanding business boundaries and penetrating supervision.

We all know that blockchain can truly record the information in the real world in the chain in a time-determined manner in a deterministic manner. Such a feature operates in an enterprise scenario, and its role is very large. It enables a downstream service to be able to perceive the historical operation of its upstream business in a cross-principal business process. If we apply these features in a business scenario, it can undoubtedly expand the boundaries of the business. If we are in the regulatory arena, it can provide a penetrating regulatory capability.

WX20191109-161455

Case: The whole process of blockchain vaccine was traced back. We use the blockchain to open up the production of vaccines, cold chain logistics, disease control centers, vaccination stations, cold storage of vaccination stations, and final vaccination and feedback. We all know that there have been some security incidents in the vaccine industry that we are not willing to see. When we went deep into this field to see this business link, we found that there are some pain points. These pain points are typical, including many business links. The opaque information and the difficulty in verifying the business process are difficult to supervise. Such a feature poses a very large security risk, thus causing security incidents that we know. The introduction of the blockchain, its value is very prominent, it can well solve the problem of information penetration and regulatory penetration.

WX20191109-161541

3. How to use the blockchain?

The value of the blockchain scene, in addition to the scene of the vaccine just can be tapped, there should be a lot of scenarios, then how should we explore more valuable scenarios? We have some simple conclusions. We think there are three points: 1. Cross-subject collaboration, 2. Multi-link, long-term transaction chain, 3. Need to establish a low-cost trust scenario. We can start from these three points to match the business areas that we can meet every day. We believe that we can find similar scenes from many business areas. But can you really find out the value points? Depending on whether we really go deep into every aspect of a business field, whether we really explore the innovation space of each link, there is no opportunity for business change, instead of putting a simple business on the chain, to talk about whether there is a blockchain. Almost, blockchain is not necessary, which may require us to go deeper.

WX20191109-161645

Second, the evolution of the new generation of blockchain architecture

We boil down to five aspects: one is the data structure, the other is the multi-network architecture, the third is the system interaction model, the fourth is the user role authorization, and the fifth is the data traffic. Behind the evolution of such a technology architecture is the change we face in new business scenarios. From these five aspects, we have implemented a blockchain underlying architecture for real-world scenarios, but we hope to implement a unified blockchain infrastructure that is compatible with public-chain scenarios and enterprise-oriented scenarios. We will introduce the five in the following. Aspects of the specific approach.

Book model. By defining standardized data representations, we have established an account model with universal data representation capabilities, and this model establishes the basis for data interoperability in different business scenarios.

Network model . We will focus on multi-book, multi-chain hierarchical network architecture. Such a layered network can have different consensus algorithms, can have different books, can have different contracts, and can establish different constraints with each other. Such a model can well correspond to the actual business business, the business needs of different levels of business, and some rules.

Interaction model . We find that the common blockchain architecture uses smart contracts as the main chain and chain interaction, but smart contracts can do some common business rule verification between consensus nodes. But in fact, we found that there is another category in the enterprise scenario. When we do some personalized out-of-band transaction processing on the chain events, we need to introduce some new architectures to implement, so we took the external trading instructions and chain. A new interactive architecture for contracts.

Permission control model . We know that in the public chain scenario, the user access rights in the chain are equal, and the public chain architecture is a peer-right permission model. But in the enterprise architecture, it is usually not equal. There may be supervisors in one model, there will be other participants, and each may need a distributed access control model with different business requirements. We solve a problem by adopting a mapping distributed permission framework of participants, users, roles, and privilege layers. In the future distributed business model, how to realize the governance ability of multi-party business ecology including supervisors.

Store the model . This is the practice from the perspective of data capacity. By discretely storing the books, it is the implementation of discrete storage that can perform hash check. It successfully breaks the block header design and breaks the design of a lightweight block header. Block size limitations, unrestricted account/transaction growth limitations, and structured/unstructured data fusion issues.

Third, the JD Chain project

WX20191109-161752

Looking back at the JD Chain project, in fact, the time to enter the engineering practice is still relatively short, only a short one year, we really started in March 2018, when Jingdong released the "Jingdong blockchain technology practice white paper 2018", to this year Just one year in March, we officially opened up JD Chain. We also deal with a lot of problems, including transaction processing performance, support for national secrets, pluggable algorithm for consensus algorithms, data-penetable retrieval, multi-password system, contract engineering and other issues.