The People's Bank of China Digital Currency Research Project Team: Advantages, Disadvantages and Development Trends of Blockchain

Introduction: Blockchain technology has high security that cannot be tampered with and high availability of heterogeneous multi-active, but it also has certain limitations in terms of performance, privacy protection, and upgrade repair.

Author: Digital Currency Research Project Team of the People's Bank of China

Article | China Finance, 2016, Issue 17

The blockchain technology originated from the Bitcoin community, and is not only valued by financial institutions, but also gradually concerned by major economies and important international organizations in the world. It is gaining momentum amid cheers and doubts. The author attempts to analyze the advantages, disadvantages and future development trends of blockchain technology, so as to make the best use of its strengths and avoid its weaknesses.

1. Advantages of blockchain

1. Cannot be tampered with, more secure

In the security solutions of traditional information systems, security relies on access control with layers of defense. Similar to a bank vault buried deep underground, high-value data has always been under the strict protection of a special computer room, a proprietary network, and a full range of security software. APIs/access interfaces are dedicated access channels opened on the iron bucket. Anyone must pass identity authentication and authorization before they can enter the database through a dedicated channel, read or write data, and leave a historical record.

There are usually two ways to protect property security: hide it so that only the owner can get it, such as gold; announce the ownership of the property to the public and endorse it with the law, such as real estate. Traditional security solutions are the first idea, and blockchain is the second. Through blockchain technology, anyone can share the database that records transactions. However, due to clever design and supplemented by cryptography and consensus algorithms, blockchain has achieved the immutability of database historical records. Practice has proved that such a database can ensure that Bitcoin, with a market value of tens of billions of dollars, can operate stably under the attacks of global hackers.

2. Heterogeneous multi-active and high availability

From the perspective of the blockchain system architecture, each system participant is a multi-active node in a different location, which far exceeds the redundancy of three centers in two locations. It is a naturally multi-active system: each full node maintains a complete copy of the data, and these data copies are under the control of different entities, and the data remains highly consistent through the consensus algorithm.

If a node encounters network problems, hardware failures, software errors, or is controlled by hackers, it will not affect the system and other participating nodes. After troubleshooting and completing data synchronization, the problematic node can be added to the system at any time to continue working. Because the normal operation of the entire system does not depend on individual nodes, each node can choose to go offline for routine system maintenance, while ensuring that the entire system works 24 hours a day, 7 days a week.

In addition, the nodes in the blockchain interact through a point-to-point communication protocol. Different nodes can be developed by different developers using different programming languages, based on different architectures, and implementing different versions of full nodes to process transactions while ensuring that the communication protocol is consistent. The resulting heterogeneous software environment ensures that even if a certain version of the software has problems, the overall blockchain network will not be affected, which is also the cornerstone of its high availability.

3. New collaboration mechanism, more efficient

For large-scale multilateral collaboration between companies, before the application of blockchain, there are usually only two solutions.

First, multiple entities are required to find a common "superior" institution, and a common trust center coordinates the entire organization. The limitation of this method is that in some scenarios, it is difficult to find a trust center that is recognized by all market participants; for a center, coordination matters must have priority, and it may not be able to meet all collaboration needs in a timely and effective manner.

Second, by jointly establishing a third-party organization, that is, all participants transfer some rights and jointly establish a third-party organization to complete the collaboration. The limitation of this method is that third-party organizations are often independent. If the system cannot meet their profit and management needs, third-party organizations can often become the actual power center of all participants. After the establishment of the third-party organization, how to absorb new members and how to adjust the roles and powers of each participant as the situation develops and changes, all rely on a lot of negotiations and transactions. Blockchain provides a different method from the traditional one: connecting participants in a peer-to-peer manner, and jointly maintaining a system by the participants, expressing the rules of collaboration through consensus mechanisms and smart contracts, and achieving a more flexible collaboration method. Because the responsibilities of the participants are clear, there is no need to transfer power to third-party organizations, and there is no need to maintain the cost of third-party trust organizations, which is conducive to better collaboration among all parties. As a trust machine, blockchain is expected to become a new collaboration model with low cost and high efficiency, forming a new collaboration mechanism with a larger scope and lower cost.

4. Smart contracts, more advanced

Smart contracts have the advantages of transparency, trust, automatic execution, and mandatory performance. However, since Nick Szabo proposed it in 1993, smart contracts have always remained at the conceptual level. The main reason is that there has long been no environment that supports the operation of trusted code, and automatic mandatory execution cannot be achieved. Blockchain makes the concept of smart contracts a reality for the first time.

In essence, a smart contract is a piece of code running on the blockchain. It is not much different from the code running on the server. The only difference is that it is more trustworthy. First of all, it is trustworthy because the code of the smart contract is transparent. For users, as long as they can access the blockchain, they can see the compiled smart contract and check and audit the code. Secondly, trustworthiness also comes from the operating environment of the smart contract. The running result of a program is not only related to the program code, but also to the data provided to the program for processing. Therefore, in addition to transparency, it is also necessary to ensure the consistency and non-tamperability of data, which is exactly the advantage of blockchain.

Therefore, once a smart contract is deployed on the blockchain, the program code and data are open and transparent, cannot be tampered with, and will be executed according to the pre-defined logic to produce the expected results. If the code-based smart contract can be recognized by the legal system, then relying on the automation advantages of the program, by combining and connecting different smart contracts to achieve different purposes, it can accelerate our move towards a more efficient business society.

2. Disadvantages of blockchain

1. Performance issues need to be resolved

Judging from the current situation, the performance problem of blockchain is mainly reflected in the contradiction between吞吐量and存储带宽.

Take Bitcoin as an example. In the public chain, the processing capacity of 7 transactions per second is far from meeting the payment needs of the entire society. At the same time, the total transaction data of Bitcoin is currently close to 80G, which is a considerable burden for the storage of ordinary computers. If the throughput is simply increased by increasing the block size, Bitcoin will soon become a system that only a few large companies can operate, which violates the original intention of decentralized design. In public chain systems such as Bitcoin and Ethereum, the above contradiction is the biggest challenge faced in system design.

In the consortium chain, because the nodes participating in bookkeeping are optional and controllable, the upper limit of the weakest node's capabilities will not be too low, and it can be improved through resource investment, and then targeted replacement of components such as consensus algorithms will ultimately achieve an all-round improvement in performance. However, the consortium chain, which is the basic support for smart contracts, has another test: when smart contracts are running, they will call each other and read and write block data, so the processing sequence of transactions is particularly important. Nodes cannot process or verify transactions in parallel, but can only do it one by one, which will restrict the processing capacity of nodes.

2. Privacy protection needs to be strengthened

Traditionally, data is stored on a central server, and the system operator protects data privacy. However, in a public chain, there is no centralized operator, and every participant can obtain a complete data backup, which also means that the public chain database is transparent.

Bitcoin's solution to privacy protection is to achieve anonymity by isolating the association between the transaction address and the real identity of the address holder. So although you can see the addresses of the sender and receiver of each transfer record, you cannot correspond to a specific person in the real world. For Bitcoin, this solution may be enough. But if the blockchain needs to carry more business, such as real-name assets, or implement specific loan contracts through smart contracts, how to store these contract information on the blockchain, and how to execute the contract without knowing the specific contract information, it is necessary to pay attention to the application progress of new cryptographic schemes such as homomorphic encryption and zero-knowledge proof on blockchain issues. In addition, by reasonably designing the data on the system chain and arranging mechanisms such as off-chain information exchange channels, some difficulties in privacy protection can also be avoided.

3. Upgrade and repair mechanisms need to be explored

Unlike the upgrade method of centralized systems, in public chains, due to the large number of nodes and the anonymity of the participants, it is impossible to shut down the system for centralized upgrades. In practice, the public chain community has explored upgrade mechanisms such as "hard fork" and "soft fork", but the remaining issues remain to be seen. In addition, since public chains cannot be "shut down", their error repair is also extremely difficult. Once a problem occurs, especially a security vulnerability, it will be very fatal.

By relaxing the restriction of decentralization, many problems can be solved. For example, in a multi-center system such as the alliance chain, upgrading the underlying blockchain by shutting down the system, or emergency intervention, rolling back data, etc., are all available means when necessary. These means help control risks and correct errors. For conventional code upgrades, by separating code and data and combining a multi-layer smart contract structure, controllable smart contract replacement can be achieved.

3. Future Trends

A new species or new phenomenon often greatly promotes the expansion of theoretical boundaries. The emergence of Bitcoin has created a new dimension of software systems. It is foreseeable that in the future, there will be a new field between the two poles of centralization and decentralization, and various blockchain systems will have different degrees of decentralization to meet the specific needs of different scenarios.

In addition to creating new blockchain systems based on new theories, it is equally important to maximize the potential of existing blockchain systems. In order to solve the performance bottleneck problem, the "Lightning Network" is a possible development direction. The "Lightning Network" moves a large number of small payments outside the main chain to form multiple payment processing centers. Through the "Lightning Network", the Bitcoin main chain sinks to the RTGS (Real-time Gross Payment System) level application, which can greatly improve the efficiency of blockchain use; "State Channel" is a more general technical idea for the "Lightning Network" outside of payment scenarios; and R3 CEV's Corda is more thorough, using blockchain only as a last resort for dispute arbitration and enforcement, playing to its strengths and avoiding its weaknesses, thereby overcoming the disadvantages of blockchain in performance, privacy and other aspects.

In the security field, although the security of the Bitcoin blockchain itself has withstood the test of many years, it still needs continuous attention. Blockchain does not solve all traditional information security problems. Blind trust in the security capabilities of blockchain may lead to serious consequences, such as the exploitation of smart contract vulnerabilities leading to the loss of digital assets. Therefore, new cryptographic schemes and key technologies in the traditional information security field need to be integrated with blockchain technology to advance hand in hand and develop in a coordinated manner. If a hybrid database combining traditional databases and blockchain is designed, different data can be processed separately and their respective advantages can be fully utilized, which is of great significance to the popularization of blockchain systems.

As more and more digital assets migrate to blockchain for cross-chain operations, interconnection between different blockchains will become inevitable. The task facing regulators is even more arduous, and they need to simultaneously consider formulating corresponding laws, regulations and technical standards to strengthen supervision and prevent risks. Can blockchain become the underlying technology of the new generation of financial infrastructure? We will wait and see. ■