A Class Of(6,4) Minimum Storage Regenerative Codes With Minimum Bandwidth For Distribution Storage Systems

With the rapid development of the Internet,the scale of data has exploded.In order to meet the rapidly growing storage demands,the distributed storage system gradually emerges.For ensuring the reliability of distributed storage system,some redundant data is usually set in the distributed storage system.These redundant data can guarantee the integrity of data in case of disk failure and other factors.In order to make the distributed storage system ensure the reliability and integrity of the stored data and enhance the overall performance of the distributed storage system,the data needs to be encoded in the system.The storage cost,repair bandwidth cost and computational complexity of the system are better than those of the uncoded system.In distributed storage system coding circuits,minimum storage regeneration code(MSR code)is one of the coding methods with good performance.Minimum storage regeneration codes not only have the characteristics of low storage overhead,security and reliability,but also have the optimal repair property of single node.In this thesis we present a(6,4)MSR code constructed in the finite field F₄.This code is reconfigurable and can correct any single node error with the minimum bandwidth,and it also has the minimum update complexity.This thesis is divided into five chapters.In the first chapter we introduces the research background and significance of distributed storage system and MSR code.In the second chapter introduces the general structure of(n=k+2,k)MSR code and its reconstruction property,the determination method of update complexity and the optimal repair property of nodes.In the third chapter,a new(6,4) MSR code is constructed in F₄,and its reconstruction property is verified and compared with the update complexity of(6,4)MSR code constructed by Sheng Guan et al.In chapter 4 verifie that(6,4)MSR codes constructed in the previous chapter can be recovered with minimum bandwidth in the case of single node failure.In the last chapter we make a brief summary of the research work of this thesis and propose some open problems to be further studied in the future.