Memory fragment treatment method, device and system

Abstract

The invention discloses a memory fragment treatment method, device and system, and relates to the technical field of computers. A specific embodiment of the method comprises the following steps: after receiving a governance request sent by any node in the distributed system when the node meets a memory fragment governance condition, judging whether a leader copy exists in the node or not; when the leader copy exists in the node, determining that each leader copy in the node corresponds to a follower copy of other nodes, and sending an election instruction to the follower copy; enabling the election instruction to instruct the follower copy and the leader copy to execute leader election, and restrict candidates of the leader election in the follower copy; and after the election result of the leader election is obtained, restarting the node. According to the embodiment, the memory fragments in the distributed system nodes can be effectively treated under the condition of not influencing services.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a memory fragment management method, device and system. Background technique [0002] In a distributed system, each node may form memory fragments due to differences in the length of stored data and frequent data updates. Memory fragmentation refers to the waste of memory space caused by the memory allocated to the process by the operating system, but the process cannot be used due to reasons such as small continuous storage space. Excessive memory fragmentation will occupy memory space, reduce process execution efficiency, and affect the performance of corresponding applications. In the prior art, two methods are mainly used to solve the problem of memory fragmentation: first, when a node is found to have excessive memory fragmentation, the node is directly restarted. Second, when the memory fragmentation of a certain node is excessive, a thread is started in the bac...

AI Technical Summary

Problems solved by technology

[0003] In the process of realizing the present invention, the inventor finds that the prior art has at least the following problems: the first method can indeed solve the problem of memory fragmentation, but if there is a leader replica (leader) stored in the restarted node, it will lead to including the The entire consistency group including the leader copy and the corresponding follower copy (follower) is unavailable, and re-electing the leader requires at least three special election cycles (special election cycles refer to the election cycles applied in fault repair scenarios), if Each special election period is 500 milliseconds, which will cause the relevant services for more than 1.5 seconds to be unavailable, which will bring great jitter to the business

The second method requires a large number of locking operations, which will have a more serious impact on the business and cannot be used in actual production systems

Images