Linux noop scheduling algorithm based read-write method and device

Abstract

The invention discloses a Linux noop scheduling algorithm based read-write method and device. The method comprises the steps that read times of all data of a queue in the preset time period are counted through a first process; data with the read times larger than or equal to a preset threshold value is stored to a cache every preset time period through a second process; read requests of to-be-read data are received; whether the to-be-read data exits in the cache or not is judged; if yes, the to-be-read data is read from the cache, and if not, the to-be-read data is read from the queue. Accordingly, data high in read frequency is moved to the cache, when the data high in frequency needs to be read, the required data can be obtained from the cache directly and does not need to be obtained from the queue, the required data is obtained from the queue only when the required data does not exist in the cache, in this way, the probability of the read requests of the noop scheduling algorithm can be lowered, and then the system read-write efficiency is improved, and the system performance is maximized.

Description

technical field [0001] The invention relates to the technical field of IO, in particular to a reading and writing method and device based on a Linux noop scheduling algorithm. Background technique [0002] At present, there are four Linux IO scheduling algorithms, namely the elevator algorithm (no operation, noop), the deadline algorithm (deadline), the anticipatory algorithm (anticipatory), and the complete fair queuing algorithm (complete fair queuing, CFQ). [0003] Among them, the noop algorithm implements a FIFO queue, which organizes IO requests like the working method of an elevator. When a new request arrives, it merges the request into the latest request to ensure that the same medium is requested. [0004] However, the noop algorithm tends to starve reading and favor writing, that is, the write request passes through the file system cache, and the next write operation can be started without waiting for one write to be completed. The write request is merged and accu...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a method and device for reading and writing based on the Linux noop scheduling algorithm, and the purpose is to solve the problem that the reading and writing efficiency of the existing Linux noop scheduling algorithm is low

Images