Distributed transaction processing system and method based on broadcast channel

Abstract

The invention provides a distributed transaction processing system based on a broadcast channel. The distributed transaction processing system comprises a main server side, a central data node, a plurality of data nodes and a plurality of terminals, the main server end is in network connection with the central data node and is in network connection with the plurality of data nodes, the terminal isin communication connection with the data node, and the central data node performs data transmission through a broadcast channel, and establishes broadcast communication connection with the main server side and the data node respectively. The distributed transaction processing method based on the broadcast channel is realized in combination with the system. According to the distributed transaction processing system and method based on the broadcast channel provided by the invention, by using the traditional database technology, the data consistency of the system is ensured, the relation of the response time of the system and the scale growth of the system is not large, and only the communication delay between nodes needs to be ensured, so that the availability of the system is excellent.A message-based consistency algorithm is not used, so that the information transmission among the nodes is reduced, and the message processing efficiency is effectively improved.

Description

technical field [0001] The present invention relates to the field of computer and communication technologies, and more specifically, to a broadcast channel-based distributed transaction processing system and a broadcast channel-based distributed transaction processing method. Background technique [0002] In a distributed system, in order to ensure that data is not lost, data is often stored in multiple copies. However, some abnormal situations will lead to data inconsistency, so consistency is a very important part of distributed systems. The data consistency problem in the distributed system is also called the distributed transaction processing problem. For distributed transaction processing, experts have proposed a variety of algorithms to solve distributed transaction processing problems. Many of the existing distributed transaction processing algorithms are based on message passing between distributed systems to achieve data consistency. Such as two-phase commit (2PC...

AI Technical Summary

Problems solved by technology

But in most cases, the distributed computing that needs to be processed often requires geographically dispersed nodes to participate together. In this case, if the distributed transaction processing algorithm of the entire distributed system is based on a message algorithm, the entire The processing process is very long, requiring a lot of computing and bandwidth resources to achieve synchronization, and the system has poor scalability, which will encounter great challenges in terms of execution efficiency and system availability

[0004] In this case, it has been proposed to adopt a divide and conquer scheme, such as: In these systems, there is a need to have data that can be layered or chunked into smaller systems in the problem to be solved, but if some data in the system It is the only data in the whole system that will be used by all subsystems, and the system layering or block processing will cause the consistency (ACID) of this type of data to be destroyed

Therefore, it may be necessary to return to the algorithm of distributed transaction processing based on message passing between distributed systems in a wide area network environment, which cannot solve technical problems.

Images