A Distributed Cache Client Based on Metadata Management

Abstract

The invention provides a memcached client side based on metadata management. The memcached client side based on metadata management comprises a metadata management module, a heartbeat service module and a load balancing module, wherein the metadata management module is used for providing metadata support and managing the metadata; the heartbeat service module is used for sending heartbeat messages to cached nodes and recording and storing response time of the heartbeat messages to the metadata; and the load balancing module is used for determining optimal cached nodes according to the cached node storage conditions stored in the metadata and the cached node response time provided by the heartbeat service module and then downloading cached data from the optimal cached nodes. By the aid of a memcached implementation method and a memcached implementation system based on metadata management, the problems of cache reconstruction, substantial decrease of hit rate and nonsupport of the client side to a copy function caused by cached node number variation or failure of some cached nodes in a memcached environment can be solved.

Description

technical field [0001] The invention relates to the computer field, in particular to a distributed cache client based on metadata management. Background technique [0002] Distributed refers to a function implemented outside the core system through a cluster. Memcached is currently the most widely used distributed cache implementation. Although Memcached is a "distributed" cache server, there is no distributed function on the server side. Each Memcached node will not communicate with each other to share information, and the support for distributed depends entirely on the implementation of the client. [0003] The current Memcache client is generally distributed through resource pools. The resource scheduling algorithms mainly include "remainder calculation method" and "consistent hash calculation method". When changing, it will cause a large area of ​​cache reorganization, seriously affecting the cache hit rate. The complexity of the "consistent hash calculation method" ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a distributed cache client based on metadata management, and implement the Memcached client in the way of metadata storage, which can solve the problem that in the distributed cache environment, when the number of cache nodes changes, or some When the cache node fails, the cache reorganization, the hit rate is greatly reduced, and the client does not support the copy function.

Images