Recovery Mechanisms In The D-sql Distributed Database Design And Realization

Distributed Database System(DDBS) is produced in the 70s of 21st century,as a result of the development of database and computer network technique.Generally speaking, DDES is a whole logically,although it is made up of many geographically separated single database system.The controllability of redundancy, by which we can control the number of the replica if need to, is an important characteristic of DDBS.for example,we can place all the replica on diferent nodes of DDBS.Although one node crashed, the system is available because of the other replica on other normal nodes. It is apparent that increasing redundancy can improve the availability of data. To ensure the consistency of database,we must update all the replica at the same time. During updating,database may be inconsistent because of some error,for example network error,so we must recover it by the recovery mechanism of the DDES.D-SQL .based on MySQL,is a highly reliable DDES.This paper introduces some important techniques to implement the D-SQL and correlative knowledge,laying emphasis on the principle of implementing the recovery mechanism.