Memory system, and method and system for controlling service quality of memory system

Abstract

The present invention provides a storage system service quality control method, system and storage system, wherein the method includes: counting the processing capability information of the hard disk in the storage system, and obtaining the processing capability of the hard disk through the processing capability information; according to the processing capability of the hard disk The size divides the cache into multiple levels, each level of cache corresponds to one or more hard disks in the storage system, and the page quota of each level of cache is pre-set Allocation is obtained; for a level cache whose dirty data reaches a preset ratio, write the data in the level cache into the hard disk corresponding to the level cache. The storage system service quality control system includes: a statistical module, a first division module and a disk writing module. The storage system includes a host, a cache, and the above-mentioned storage system service quality control system. The invention overcomes the preemption phenomenon of internal page resources in the cache.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a storage system service quality control method, system and storage system. Background technique [0002] Cache (cache) is a component introduced by the storage system to improve performance. It is mainly used to improve the input / output (Input / Output; hereinafter referred to as: I / O) performance of the entire system, and stores some temporary data in the memory to speed up The host's I / O response. For the read operation of the host, the system temporarily puts the read data into the cache, and the read operation of the host directly hits the cache, thereby saving the operation of the host to read the disk; for the write operation of the host, the system directly stores the write data of the host Cached in the cache, thus saving disk write operations. In order to make the read and write operations go as smoothly as possible, the cache will read the data that the host ...

AI Technical Summary

Problems solved by technology

However, page resource preemption still occurs inside each cache partition in the prior art, resulting in the performance of the business corresponding to this cache partition (such as transmission delay, packet loss rate, etc.) being dragged down by the slowest disk at the back end. Thereby reducing the quality of service of the storage system

Images