Multi-level cache method based on historical upgrading and downgrading frequency

Abstract

The invention provides a multi-lever cache method based on historical upgrading and downgrading frequency. The multi-lever cache method based on the historical upgrading and downgrading frequency is based on historical hinting information of a data block, and the historical hinting information is one of the key points of a multi-lever cache system. Through the hinting frequency, a hot data block can be identified effectively, the hot data block can be stored in the cache at a higher lever for a longer time, the data block hit rate in the system is increased, and the average response time is reduced; according to the multi-lever cache method based on the historical upgrading and downgrading frequency, a traditional LRU stack is divided into two special queues, so that the hinting information has locality, mixing of the hot data block and a cool data block is avoided, and the use of band width between the caches in all levers; the hot data block is identified effectively, the hot data block can be stored in the cache at the higher lever for a long time through an algorithm, the downgrading and upgrading operation of the caches in all levers is reduced, and the band width consumption between the caches is reduced further; the space consumption is small, and basis is provided for a better read-write property of the system under different loads.

Description

technical field [0001] The invention relates to the field of data reading and writing and storage of computer systems, in particular to a multi-level cache method based on historical de-escalation and de-escalation frequencies. Background technique [0002] In a large data center, heterogeneous storage devices work together to accelerate data read and write operations. Characteristically, the high-level storage device acts as a cache for the low-level storage device, forming a distributed multi-level cache system. In recent years, multi-level cache systems have received increasing attention due to their high I / O performance, low monetary cost, and high flexibility. [0003] In the past two decades, many typical multi-level caching solutions have been proposed to improve the I / O performance of storage systems. One of the most effective methods is to build another cache between different levels, and use implicit identification of hot data blocks in this cache. These hints p...

AI Technical Summary

Problems solved by technology

[0005] However, existing multi-level caching algorithms still have potential for improvement

On the one hand, most multi-level caching algorithms use implicitly stored centralized information of data blocks, but valuable historical implicit information is ignored; on the other hand, some algorithms only save the implicit information of the last few operations, and does not adequately describe the state of the data block

Images