The Design Of Data Coding And Fault Tolerant Recovery Tool Based On RS Erasure Code

Nowadays,with the progress of information science,storage system is required to have the characteristics of low cost and high reliability.Erasure Code technology can well solve the problem,but the existing Erasure Code tools(such as Jerasure,etc.)implemented by Erasure Code technology are mainly run on Linux platform.The few Erasure Code tools that support Windows platform(such as ISA-L)are mostly implemented with C language,and the environment configuration is complex,which brings some limitations to Windows users.Therefore,it is very meaningful to design and implement an Erasure Code tool that can be used in Windows platform and the environment is easy to configure.This paper designs and implements a data coding and fault-tolerant recovery tool which is based on RS Erasure Code and can run on both Windows platform and Linux platform.This tool adopts Python language development,which has the advantages of fast speed,cross-platform,simple operation and easy environment configuration.This paper first describes the construction of Galois Field and the definition of four operations in this field.Then,Vandermonde RS Erasure Code technology is used to realize the encoding and decoding of this tool.As well as,the Num Py and Numba libraries are used to accelerate the running of the program,which solves the problem of low execution efficiency of Python programs.Finally,this tool is tested.In this paper,the tool was tested on Linux and Windows platforms respectively.The test content included time consumption,CPU occupation rate and memory occupation rate.First,on the same Linux platform,the tool and Jerasure library were tested,and the two were compared in terms of time consumption,CPU occupation rate and memory occupation rate.The experimental results show that the encoding and decoding speed of this tool is faster than Jerasure library.The CPU resource occupation of this tool is similar to that of Jerasure library,while the memory resource occupation of this tool is slightly larger than that of Jerasure library,but the difference is small and can be ignored to a certain extent.After that,the tool was tested on the Windows platform,and the test content was the same as that of the tool on the Linux platform.The experimental results show that the encoding and decoding speed of this tool on Linux platform and Windows platform are equivalent.This tool's CPU,memory resource occupation is stable.