CN111371850A

CN111371850A - Multi-partition PBFT (partition-based multi-partition function) based multi-channel block chain platform optimization method

Info

Publication number: CN111371850A
Application number: CN202010109423.2A
Authority: CN
Inventors: 胡建国; 谭德志; 丁颜玉
Original assignee: Guangzhou Intelligent City Development Institute; National Sun Yat Sen University
Current assignee: Guangzhou Intelligent City Development Institute; National Sun Yat Sen University
Priority date: 2020-02-22
Filing date: 2020-02-22
Publication date: 2020-07-03
Anticipated expiration: 2040-02-22
Also published as: CN111371850B

Abstract

The invention discloses a multi-partition PBFT (PBFT) based multi-channel block chain platform optimization method, and relates to a block chain technology. The method is applied to a block chain network based on a Fabricius 1.0 multichannel architecture, a plurality of partitions with the same running number are constructed on each Peer node, each Peer node runs a corresponding PBFT consensus protocol on different partitions, in the same PBFT consensus protocol, a partition backup relation capable of responding is established between the corresponding partition on one Peer node and the corresponding partitions on other Peer nodes, and the distributed consistency process of optimizing logs is executed by adopting a two-segment PBFT consensus algorithm. The invention introduces the PBFT algorithm after the transformation to solve the Byzantine fault tolerance problem existing in the original design scheme, transforms the common identification node into the common identification node with multiple partitions, and realizes the PBFT parallel queues in different partitions, thereby realizing the data isolation and improving the concurrency.

Description

Multi-partition PBFT (partition-based multi-partition function) based multi-channel block chain platform optimization method

Technical Field

The invention relates to the technical field of block chains, in particular to a multi-partition PBFT-based multi-channel block chain platform optimization method.

Background

With the key concern of bitcoin in the world, a large number of experts and scholars develop research on underlying zone block chain technology, and more financial technology companies lay out zone blocks chains. In 2015, 12 months, IBM and other companies announced a federation to develop an open-source federation chain platform, hyper-ridger Fabric, in order to better advance the development of blockchain technology. The platform realizes the transition from the experimental stage to the release stage, is different from the platforms of public chains such as bitcoin, Ethenfany and the like, and the Hyperhedger Fabric maintains the state and the chain code of the Docker container through application coordination service. In order to achieve the token-free characteristic, a PBFT (physical Byzantine Fault Tolerance) algorithm is introduced as a consensus algorithm, and the algorithm has certain advantages in terms of system throughput, transaction confirmation time and the like.

After HyperLegger Fabric1.0, IBM recombines the architecture, decouples the single node model of the original 0.6 version into three node types of a client node, a Peer node and a sequencing service node, introduces Kafka consensus and Solo consensus, removes the original PBFT consensus algorithm, and has another important characteristic of introducing multiple channels for data isolation.

However, the existing Kafka consensus does not have the property of Byzantine fault tolerance, the Kafka mechanism is only applicable to the scene of master-slave preparation and not applicable to the scene of Byzantine fault tolerance, and once the leader consensus node goes bad, wrong information can be diffused to the whole distributed network. And the native PBFT algorithm is not suitable for the novel HyperLegger Fabric architecture. The native PBFT has three sections and finally executes instructions on the node and returns them to the client, whereas in the blockchain scenario, the consensus algorithm works better to perform distributed synchronization of the logs without executing them. Meanwhile, the native PBFT adopts a single-view mode to synchronize data and elect a master node, but this mode is only applicable to a single-channel case, and in consideration of the multi-view characteristic introduced in fabric1.0, the native PBFT needs to be modified so that multiple channels form multiple parallel queues to improve the concurrency of the system.

Disclosure of Invention

In view of the shortcomings of the prior art, the present invention aims to provide a multi-partition PBFT-based multi-channel blockchain platform optimization method.

In order to achieve the purpose, the technical scheme adopted by the invention is as follows:

a multi-partition PBFT-based multi-channel block chain platform optimization method is applied to a block chain network based on a Fabrico 1.0 multi-channel architecture, the blockchain network includes client nodes, a ranking service, and Peer nodes, a multi-channel support service provided based on the ranking service, said client node and said Peer node being connected to a given channel and sending and receiving messages over the given channel, wherein, a plurality of partitions with the same running number are constructed on each Peer node, each Peer node simultaneously runs a corresponding PBFT consensus protocol on different partitions, in the same PBFT consensus protocol, and a corresponding partition on one Peer node and corresponding partitions on other Peer nodes establish a responsive partition backup relationship, and the distributed consistent process of the optimized log is executed by adopting a two-stage PBFT consensus algorithm.

Preferably, in the multi-partition PBFT-based multi-channel block chain platform optimization method, one partition corresponds to one channel in multiple channels, data in each channel is independent of each other, and in the same PBFT consensus protocol, a corresponding partition on one Peer node and corresponding partitions on other Peer nodes form a logical cluster together.

Preferably, in the multi-partition PBFT-based multi-channel block chain platform optimization method, partitions on Peer nodes in each PBFT consensus protocol are created as partition nodes in a corresponding PBFT consensus protocol, where the partition nodes include a master partition node and slave partition nodes, and the number of partitions on the same Peer node that can be created as the master partition node is at most 1.

Preferably, in the above method for optimizing a multi-partition PBFT-based multi-channel block chain platform, the step of executing the two-segment PBFT consensus algorithm includes: .

S1, the primary partition node sends the ordering request received from the client node to

All the slave partition nodes attach the signature of the master partition node on the message and the sequencing result of the user instruction when sending the message;

s2, after receiving the broadcast ordering request and confirming, the slave partition nodes intercommunicate with each other

And sending confirmation to the request, and when the main partition node receives confirmation of more than half of other slave partition nodes to a certain request, regarding the request as consensus in the cluster, and storing the information in a self log.

The invention has the beneficial effects that: the optimization method introduces the modified PBFT algorithm, can solve the Byzantine fault tolerance problem of the original PBFT algorithm in the Hyperhedger Fabrict 1.0 architecture, simultaneously, the communication complexity can be reduced from O (2n ^2) to O (n ^2) by adopting the two-stage PBFT consensus algorithm in the distributed consensus process of the optimized log, the consensus can still be successfully achieved in the cluster consisting of malicious nodes with less than one third of the total node number, the malicious information can not enter the consensus log, and the transaction speed can also be improved. The PBFT protocol maintained by each different partition uses the common identification means, so that a plurality of parallel channels can be identified, and the concurrency is increased. Meanwhile, the multiple channels enable a given peer node set to receive blocks containing related transactions, so that complete isolation is achieved from other transactions, and data isolation and confidentiality are achieved.

Drawings

FIG. 1 is a block diagram of a multi-channel configuration of the present invention;

FIG. 2 is a schematic structural diagram of a multi-partition consensus node according to the present invention;

FIG. 3 is an information flow diagram of the two-segment PBFT consensus algorithm of the present invention.

Detailed Description

The invention will be further described with reference to the accompanying drawings and specific embodiments thereof, it being understood that the invention is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the following drawings before any embodiments of the invention are explained in detail. The invention is capable of other embodiments and of being practiced or of being carried out in various ways. Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of "including," "comprising," or "having" and variations thereof herein is meant to encompass the items listed thereafter and equivalents thereof as well as additional items. The terms "mounted," "connected," and "coupled" are used broadly and encompass both direct and indirect mountings, connections, and couplings. Further, "connected" and "coupled" are not restricted to physical or mechanical connections or couplings, but may include direct or indirect electrical or electrical connections.

It should be noted that the present invention may be implemented using a plurality of hardware and software based devices as well as a plurality of different structural components. Furthermore, and as described in subsequent paragraphs, the specific configurations illustrated in the drawings are intended to exemplify embodiments of the invention and that other alternative configurations are possible. Unless otherwise specified, the terms "processor," "central processing unit," and "CPU" are interchangeable. Where the term "processor" or "central processing unit" or "CPU" is used to identify a unit that performs a particular function, it should be understood that unless otherwise noted, these functions may be performed by a single processor or by multiple processors (arranged in any form, including parallel processors, serial processors, or cloud processing/cloud computing configurations).

As shown in fig. 1 to fig. 3, an embodiment of the present invention provides a multi-partition PBFT-based multi-channel blockchain platform optimization method, which is applied to a blockchain network based on a fabric1.0 multi-channel architecture. The blockchain network comprises client nodes, sequencing services and Peer nodes, wherein the client nodes and the Peer nodes are connected to a given channel and send and receive messages through the given channel based on the multi-channel support services provided by the sequencing services.

Specifically, the sequencing service provides a shared channel for the client node and the Peer node to realize a broadcast service of a transaction, when a network is started and configured, the Peer node and the client node which need to be connected with the channel are registered on the channel, a message broadcasted on the channel is finally sent to all the Peer nodes in the channel through atomic broadcast, and the sequencing service is responsible for distributing the messages on different channels to the corresponding channels.

The native PBFT provides that only one PBFT protocol can be operated in one cluster, and one physical node can only serve as one node type, and the mode is only suitable for the case of a single channel by adopting a single-view mode to synchronize data and elect a main node. To do so, the native PBFT needs to be modified so that multiple channels form multiple parallel queues to increase the amount of concurrency of the system. As an improvement of the invention, a plurality of partitions with the same running number are constructed on each Peer node, and each Peer node simultaneously runs a corresponding PBFT consensus protocol on different partitions. In the same PBFT consensus protocol, a corresponding partition on one Peer node and corresponding partitions on other Peer nodes establish a partition backup relationship which can respond. In order to reduce the communication complexity and improve the transaction speed, the distributed consistent process of optimizing the log is executed by adopting a two-section PBFT consensus algorithm.

Specifically, in a preferred embodiment of the present invention, one partition corresponds to one channel in multiple channels, data in each channel are independent of each other, and in the same PBFT consensus protocol, a partition corresponding to one Peer node and partitions corresponding to other Peer nodes form a logical cluster together. The partitions on the Peer nodes in each PBFT consensus protocol are created into partition nodes in the corresponding PBFT consensus protocol, the partition nodes comprise main partition nodes and slave partition nodes, and the number of the partitions on the same Peer node which can be created into the main partition nodes at most is 1. Therefore, a plurality of PBFT consensus protocols can be simultaneously operated on a single Peer node, and each partition corresponds to one PBFT consensus protocol. In order to improve the concurrency quantity, the main partition node of each PBFT consensus protocol does not fall on the same Peer node.

Specifically, as shown in fig. 2, a case where 3 PBFT consensus protocols are simultaneously run on four Peer nodes is shown. As shown in fig. 2, three partitions run on each Peer node, which are respectively represented by circles "1", "2", and "3", and partitions "1" on four

Peer nodes

1, 2, 3, and 4 jointly form a PBFT consensus protocol, and so on, each partition forms a logical cluster. Each logic cluster runs a respective PBFT consensus protocol, and the main partition nodes of the three logic clusters are not concentrated on one Peer node and respectively respond to user requests in parallel.

Further, in a preferred embodiment of the present invention, as shown in fig. 3, the two-segment PBFT consensus algorithm is performed by:

s1, the master partition node sends the ordering request received from the client node to all the slave nodes

The partition node attaches the signature of the main partition node on the message and the sequencing result of the user instruction when sending the message;

s2, after receiving the broadcast ordering request from the partition node and confirming, communicating with each other,

and sending confirmation to the request, and when the main partition node receives the confirmation of more than half of other slave partition nodes to a certain request, regarding the request as achieving consensus in the cluster, and storing the information in a self log.

Since only the synchronization of the data is considered and the execution of the data is not required to be considered, only two-step consensus is needed to achieve the effect of data synchronization.

In conclusion, the optimization method introduces the modified PBFT algorithm, which can solve the problem of Byzantine fault tolerance of the original PBFT algorithm in the Hyperhedger Fabrict 1.0 architecture, and meanwhile, the communication complexity can be reduced from O (2n ^2) to O (n ^2) by adopting the two-section PBFT consensus algorithm to execute in the distributed consensus process of optimizing the log, but the consensus can still be successfully achieved in the cluster formed by malicious nodes with less than one third of the total node number, so that the malicious information can not enter the consensus log, and the transaction speed can also be improved. The PBFT protocol maintained by each different partition uses the common identification means, so that a plurality of parallel channels can be identified, and the concurrency is increased. Meanwhile, the multiple channels enable a given peer node set to receive blocks containing related transactions, so that complete isolation is achieved from other transactions, and data isolation and confidentiality are achieved.

The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are merely illustrative of the principles of the invention, but that various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined by the appended claims and their equivalents.

Claims

1. A multi-partition PBFT-based multi-channel block chain platform optimization method is applied to a block chain network based on a Fabrico 1.0 multi-channel architecture, the blockchain network includes client nodes, a ranking service, and Peer nodes, a multi-channel support service provided based on the ranking service, said client node and said Peer node being connected to a given channel and sending and receiving messages over the given channel, the method is characterized in that a plurality of partitions with the same running number are constructed on each Peer node, each Peer node simultaneously runs a corresponding PBFT consensus protocol on different partitions, in the same PBFT consensus protocol, and a corresponding partition on one Peer node and corresponding partitions on other Peer nodes establish a responsive partition backup relationship, and the distributed consistent process of the optimized log is executed by adopting a two-stage PBFT consensus algorithm.

2. The multi-partition PBFT-based multi-channel block chain platform optimization method of claim 1, wherein one partition corresponds to one channel in a multi-channel, data in each channel are independent of each other, and in the same PBFT consensus protocol, a corresponding partition on one Peer node and corresponding partitions on other Peer nodes jointly form a logical cluster.

3. The multi-partition PBFT-based multi-channel blockchain platform optimization method according to claim 1, wherein partitions on Peer nodes in each PBFT consensus protocol are created as partition nodes in a corresponding PBFT consensus protocol, the partition nodes include a master partition node and slave partition nodes, and the number of partitions on the same Peer node that can be created as the master partition node is at most 1.

4. The multi-partition PBFT based multi-channel blockchain platform optimization method according to claim 3, wherein the execution steps of the two-segment PBFT consensus algorithm include: