Method and system of database management for replica database

Abstract

When a database area multi-written or replicated in an external storage unit is separated, a database management system writes consistent data into the external storage unit. A method and system capable of controlling a timing of separating the replicated database area is provided. Furthermore, new data directory information necessary to access the replicated database is dynamically added based on the data directory information of the replication source database. A method and system capable of selecting whether to access the database which is the source of the replication or access the replicated database are provided. Thus, when the database area multi-written or replicated in an external storage unit is separated and physical replicated databases are created, it is possible to maintain consistency of the replicated databases while continuing transaction processing.

1 . A data processing unit, wherein 
 in a database management system in an external storage unit provided with multiple storage areas for storing data, a multiple write mechanism that allows multiple write of same data, a multiple write cancellation mechanism that separates multi-written areas at an arbitrary time to allow independent reading/writing as different control areas, and a physical copy mechanism that allows mutual copying of data stored in respective areas, the data processing unit comprises:    first data synchronizing means for writing, into said external storage unit at once, unwritten data in a data cache area in a main storage unit and update data by all transactions in progress, and for delaying the write to the external storage by temporarily prohibiting or putting off an update by a newly generated transaction, or by carrying out the update only in the data cache;    first data synchronization canceling means for restarting an update of the data under control of the first synchronizing means;    second data synchronizing means for erasing all data in the data cache area in the main storage unit when all transactions in progress are completed, and for temporarily prohibiting or putting off access by a newly generated transaction in meantime; and    second data synchronization canceling means for restarting data access under control of the second synchronizing means.    
   
   
       2 . A data processing unit, wherein 
 in a data management system in an external storage unit provided with multiple storage areas for storing data, a multiple write mechanism that allows multiple write of same data, a multiple write cancellation mechanism that separates multi-written areas at an arbitrary time to allow independent reading/writing as different control areas, and a physical copy mechanism that allows mutual copying of data stored in respective areas, the data processing unit comprises:    data synchronizing means; and    data synchronization canceling means,    wherein the data synchronizing means includes:    a first status which allows both references and updates of all transactions in an operation of input/output to/from the external storage unit through a data cache area in a main storage unit in response to a data access request from a transaction;    a second status which allows references of all transactions and updates of a transaction in progress, and prohibits an update of a newly generated transaction;    a third status which allows references of all transactions and updates of a transaction in progress, and holds an update of a newly generated transaction;    a fourth status which allows references of all transactions, and allows updates of all transactions only in a cache;    a fifth status which allows a reference and/or update of a transaction in progress, and prohibits a reference and/or update of a newly generated transaction; and    a sixth status which allows a reference and/or update of a transaction in progress, and holds a reference and/or update of a newly generated transaction, and    the data synchronizing means allows in the respective statuses,    a first transition for changing from the first status to the second or third status;    a second transition whereby upon completion of all transactions in progress in the second or third status, all unwritten data in the data cache at that point in time are written into the external storage unit at once, and the status is changed to the fourth or first status;    a third transition for changing from the fourth status to the first status;    a fourth transition for changing from the first status to the fifth or sixth status; and    a fifth transition whereby upon completion of all transactions in progress in the fifth or sixth status, all data in the data cache at that point in time are erased from the data cache, and the status is changed to the first status.    
   
   
       3 . A data processing unit, wherein 
 in a database management system in an external storage unit provided with multiple storage areas for storing data, a multiple write mechanism that allows multiple write of same data, a multiple write cancellation mechanism that separates multi-written areas at an arbitrary time to allow independent reading/writing as different control areas, and a physical copy mechanism that allows mutual copying of data stored in respective areas, the data processing unit comprising:    data directory information dynamic copying means capable of copying, based on data directory information registered in the database management system to access the storage area in which a database is stored, common information independent of an access path of the storage area such as a name of the database and structural information, rewriting only information of the access path to the storage area to information of access to another arbitrary storage area, and dynamically adding second data directory information with a new name storing a name of a first data directory information of a copy source;    data directory information dynamic registering means capable of registering a newly created second data directory information in the database management system or another database management system; and    access target allocating means for changing, when the name of the first data directory information is stored in a second data directory information, a database access request from an application program to be executed according to the first data directory information so that the database access request is executed according to the second data directory information.    
   
   
       4 . The data processing unit according to  claim 3 , further comprising: 
 data directory information dynamic copying means capable of adding an arbitrary generation identifier to the newly added second data directory information;    access target allocating means for selecting one or more data directory information groups by a generation identifier or a name of the newly added second data directory information to allocate an access target databases;    access target setting means capable of specifying the generation identifier or a name of a newly added data directory information in units of clients accessing the database; and    current access generation setting means for determining all access requests from a client, for which no generation identifier or no name of the newly added data directory information is set, as access requests for which a specific generation identifier is uniformly specified.    
   
   
       5 . A database access method used in a data processing system that creates another application specific database from one or more databases used in an on-line processing application to process individual applications, wherein 
 a processing step of creating replicated databases from an arbitrary master database used in the on-line processing application is combined with a processing step of creating an application specific database from the replicated databases into single batch processing.    
   
   
       6 . A data analysis method used in a data processing system that creates another application specific database from one or more databases used in an on-line processing application to process individual applications, comprising 
 a processing step of creating replicated databases from an arbitrary master database used in the on-line processing application,    wherein the replicated databases, which store related detailed data, based on a counting result using the replicated databases are accessed by drill through.