Research And Implementation Of Host-Base CDP Key Technology Based On Linux File

As an emerging technology in the field of data storage and disaster recovery, continuous data protection (CDP) has a wide range of application prospect in the study of information system survivability. The implementation of CDP technology can diminish the loss of protected system's key data and the time cost happened from disaster to data recovery for the system that have high requestment of the business continuity and reliability demands, so as to provide arbitrary recovery point objective and minimal recovery time objective for enhancing the survivability of protected system. This article has been gotten the support of 863 project'Hologram mechanism based network survivability system model and it's key technology research'. The principal work and research results include the following four aspects.Firstly, the possible solutions, implementation methods and the future development trends of CDP technology have been studied. At the same time, the implementing mechanisms and working processes of Ext3 journaling file system have been analyzed. Further more, the principle and advantage and disadvantage of current chief snapshot technology have been stated. All of these have laid down a solid theoretical foundation to build a host-base CDP system based on Linux file, which is based on the comprehensive use of Ext3fs journaling technology and software snapshot technology based on file.Secondly, a continuous data protection model based on file has been raised over Ext3 journaling file system platform. According to this model, a host-base CDP system framework based on file has been designed. The functional modules and data flowing process of the framework have been specified. What's more, some solution methods of implementing the CDP system's key links have been given.Thirdly, on the basis of software snapshot algorithm based on file, core algorithm of the host-base CDP system based on file has been studied and designed. Especially, the snapshot version-information has been involved into the CDP algorithm to achieve the synthetic data reference schema. Based on the researches above, write operation capture between two snapshots at continuous point in time and data recovery at any point in time have been realized.Finally, taking the additional operation of data block for example, a prototype of local host-base CDP system based on file over Linux platform has been done. And the results have demonstrated that the architecture and core algorithm of the prototype system can work well. By analyzing the performance and efficiency of the prototype system, the preliminary results are showed as follow: when implementing continuous data protection at block level, the CDP prototype system is more stronger on data capture at continuous point in time and data recovery at any point in time, and it shows stable performance and greater advantage, which meets the demands that the service roaming system model asks for the data survivability.This research is an important part of the national 863 project, which has a well prospect of being applied in the document-flow system of Chinese Armed Police Force.