Research Of Storage Code Base On Binary

The explosive growth of data brings opportunities and challenges to the storage industry,the most important of these is how to ensure the reliability of stored data.The erasure code-based coding technology is widely used in storage systems because it can effectively reduce the storage overhead of the storage system while ensuring the reliability of the storage system.The main research directions are: improve coding efficiency,reduce storage overhead,and reduce decoded data volume.The main goal of the paper is to improve the coding and decoding efficiency of the erasure code,reduce storage overhead and and reduce decoded data volume,start with Zigzag decodable code that encodes based on shift and XOR operations,and use Zigzag decoding to get source data,conducted the following research works: 1.Zigzag decodable code based on cyclic symmetric matrix is proposed.Zigzag ecodable code is a type of binary code that is encoded based on shift and XOR perations and decoded using a Zigzag decoding algorithm.The computational complexity is low,but in the existing Zigzag decodable code scheme,there is a problem that the parity data block is not equal in length and the number of constructed parity data block is limited.Taking this as a starting point,this paper proposes a Zigzag decodable code based on cyclic symmetric matrix.Compared ith the existing Zigzag decodable code,the advantage is that the constructed parity data block is equal in length,which reduces the storage management overhead and storage space consumption.The number of paity data blocks is increased,which improves the fault tolerance of the storage system.2.An improved Zigzag decoding algorithm is proposed.Compared with the Zigzag ecoding algorithm that recovers only one bit of original data from the parity data lock per decoding,the improved Zigzag decoding algorithm maximizes the mount of original data recovered from the decoded data block each time,and the algorithm has a wide range of applications.The decoding efficiency is higher than the Zigzag decoding algorithm.3.A new In-place decoding scheme is proposed,which is suitable for Zigzag decodable code based on cyclic symmetric matrix.When decoding is performed using the Zigzag decoding algorithm,the amount of decoded data required to participate in the decoding of the decoded data block is different.In the new In-place decoding scheme,each parity data block needs to intercept the data of the original data block size and transmit it to the data collector to complete the decoding.In the decoding process,even if different parity data blocks are selected to participate in decoding,the amount of decoded data is also the same,and the original data decoded and restored in the decoding process uses the same storage space as the verification data participating in the decoding,which reduces the consumption of the storage space.