Sending method and apparatus, and obtaining method and apparatus of lease data based on distributed applications

Abstract

The invention discloses a sending method and apparatus, and an obtaining method and apparatus of lease data based on distributed applications. The sending method includes: a buffer structure with a lease mechanism is established; a registration request of a server is received, the server which satisfies a registration condition is selected as a main server, the lease data in the lease mechanism is sent to the main server, and the main server is authorized to provide data service; and a renewal request of the main server is received, whether the main server satisfies a renewal condition is determined, if yes, the renewal is successful, and the data service is continuously provided; and if not, the renewal is failed, and the step for receiving the registration request of the server is entered. Therefore, according to the method, the execution service function is strong, the server realizes the hot standby function during multi-unit automatic operation, based on low cost and high time efficiency, a standby unit can take over the service in a second level when one server has faults, business loss brought by the single-point problem is reduced, problems of single-point faults and split brain are thoroughly solved, and artificial intervention is not needed in the whole process.

Description

Technical field [0001] The present invention relates to the field of Internet technology, and in particular to a method for sending and acquiring lease data based on distributed applications, as well as a sending device and an acquiring device. Background technique [0002] In distributed application scenarios, there are more complex application scenarios, such as resource scheduling with functions such as red envelopes, advertising volume distribution, or monetary allocation. Usually for resource scheduling with such functions, existing methods such as rewriting the mysql engine, increasing the hardware configuration of the distribution machine, and semi-automatic hot backup are used in the field. However, these methods are too heavyweight and not only costly, but also require manual intervention. . [0003] The above resource scheduling is usually under the condition that the entire network service needs to consume the same piece of data at the same time. In order to ensure the ...

AI Technical Summary

Problems solved by technology

Usually for resource scheduling with such functions, methods such as rewriting the mysql engine, increasing the hardware configuration of the distribution machine, and semi-automatic hot backup are currently used in the field. However, these methods are too heavyweight, not only cost too much, but also require manual intervention.

[0003] The above resource scheduling is usually under the condition that the entire network service needs to consume the same data at the same time. In order to ensure user physical examination and service performance, a piece of data will be divided into multiple pieces and distributed to many servers for consumption at the same time; The server is still very important, because there is only one copy of data, which is prone to single point of failure or split-brain problem under simultaneous service of multiple machines

Once the server responsible for distribution fails to provide services normally, it will affect the user experience in the entire distribution application scenario and damage the business interests

Images