Research On Efficient SQLite Data Recovering Technology Based On MapReduce

Data recovery technology is a technical means to recovery deleted or destroyed data to normal data.It is widely used in scenarios where data is accidentally deleted or corrupted,as well as digital forensics.However,with the gradual increase in the capacity of storage devices,the slow speed of data recovery is increasingly becoming a prominent problem at present.Efficient data recovery has become a hot and difficult issue in the field of data security and digital forensics.In this paper,taking the SQLite database as an example,the efficient data recovery technology based on parallel computing model Map Reduce is deeply studied.Firstly,a method to improve the speed of SQLite data recovery based on file format by using Map Reduce is proposed.The recovery method designs the operation of page traversal,content analysis and data recovery in the traditional SQLite database recovery process as a separate Map process,while the recovered page data collation and integration operation is designed as a Reduce process.And using parallel computing mode of Map Reduce to speed up data recovery.The experimental results show that the data recovery speed of the proposed method is 2.7 times faster than that of a single server when three servers are used.Secondly,a method to improve the speed of SQLite data recovery based on WriteAhead Logging by using Map Reduce is proposed.In this method,the Write-Ahead Logging characteristic in the SQLite database is used to recover the deleted WriteAhead Logging file from the image of the file system.Then the Write-Ahead Logging is parsed to recover the data previously written to the cache to recover the historical data in the database.In practice,the first step to restore the Write-Ahead Logging file from the image of the file system will take a lot of time,and using the Map Reduce parallel computing model to speed up the operation can greatly improve the performance.The evaluation results show that when three servers are used for data recovery operations,the speed is about twice as fast as using only a single server.In this work,the existing SQLite database recovery algorithms are logically analyzed and functional partitioned,and the parallel computing model Map Reduce is used to improve their performance,which improves the recovery speed and efficiency of these algorithms.The research results of this paper show that it is feasible to use the parallel computing model Map Reduce to improve the efficiency of data recovery.However,how to effectively utilize the computing resources of large-scale service clusters still needs further research.