A cache data access method and data cache system

Abstract

The invention discloses a cache data access method and a data cache system. The method comprises the following steps of receiving an access request of a user, wherein the access request carries key information; searching a hash table to obtain a corresponding first internal memory block according to the key information; transferring the first internal memory block to a global line replaceable unit (LRU) and a first linked list head part with the grade being identical to that of the written data in the user LRU of the corresponding user if the access request is the request for writing data, wherein an internal memory block linked list affiliated by all users is preset in the user LRU; taking the internal memory block staying at the tail part of the first linked list to be used for storing the written data if the occupation of the internal memory reaches the internal memory limit and the first linked list is not empty; otherwise, acquiring a second linked list with the grade being identical to that of the written data in the global LRU, and taking the internal memory block on the tail part of the second linked list to be used for storing the written data; and transferring the internal memory block with the written data to the head part of the first linked list and the second linked list.

Description

technical field [0001] The present invention relates to the technical field of data storage, and more specifically, to a cache data access method and a data cache system. Background technique [0002] Memcached is a distributed memory key / value cache system. "Distributed" means that the cache can be distributed to multiple machines, "memory" means that all cached content is stored in memory, and "key / value" means that the cached content is the key (key) and the corresponding value (value). [0003] In a single-user environment, the common caching method is to set up Memcached by the user himself, specify the available memory size and monitoring address of Memcached with command line parameters, and then use the program to access the cached content through the corresponding address through the Memcached client library. If the size of a single Memcached is not enough, set up multiple Memcached, and then use the built-in distributed algorithm of the Memcached client library to...

AI Technical Summary

Problems solved by technology

[0005] Through the research of the inventor, it is found that in the existing Memcached application, users generally do not use up the Memcached quota allocated to them. After multiple data addition, deletion and modification requests, the memory pool will receive multiple requests to allocate or release memory blocks, so that the memory usage of Memcached in the operating system will still reach the upper limit allocated at startup, so the memory utilization rate is not high

Especially in the cloud computing environment, there are a lot of users. If you still set up a separate set of Memcached for each user, there will be problems of high management difficulty and low memory usage

Images