A Key Value Storage Data Placement Strategy For Different Types Of Load

Key-value storage(Key-Value Storage)is a new type of storage that has been widely adopted in the field of encrypted storage in recent years.Key-value storage adds the concept of Key to the original data transmission that when the data is accessed,the Key-value corresponding to the data needs to be obtained,and then the physical address and offset of the data are determined by the key value,and finally the data can be accessed.Compared with the original data transmission and access forms,key-value storage is more accurate and confidential,and has been widely used in many models and practical applications.This article comprehensively considers the characteristics of key-value storage.On the basis of the original key-value storage,key-value data transmission and access are further optimized,the load model is constructed,and the effect of optimization is tested in the actual load,including :(1)Sort and divide key-value data according to the access frequency,and improve the efficiency of the read-write access of the key-value data as a whole(2)Adjust the location of the data according to the load type to enable the CPU to execute more instructions in the same time.(3)On the basis of the above work,we propose a new key-value storage strategy KRSS(Key-Rank Storage Strategy),and verify the effectiveness of the strategy through a large number of control experiments in Level DB.(4)Summarize and evaluate the performance of KRSS when facing various types of loads,and summarize and review the advantages and disadvantages when facing actual scenarios.The new key-value storage strategy based on access popularity and load type proposed in this paper can provide a reference for the storage and access of Key-Value data in various media.KRSS can be applied to a variety of different loads,and brings significant performance optimization and improvement in data reading and writing.