[Reading notes] ORACLE11GR2 Dataguard Architecture

First, Data Guard Architecture Diagram

650) this.width=650; "Src=" Http://s1.51cto.com/wyfs02/M01/82/E4/wKioL1djv5vBZnnQAAEbqRijW5A505.png-wh_500x0-wm_3 -wmp_4-s_2983711765.png "title=" DG Architecture diagram (1). png "alt=" wkiol1djv5vbznnqaaebqrijw5a505.png-wh_50 "/>

Figure 1.1 Data Guard architecture Diagram

Data Guard( DG) consists of 5 parts, namely Redo data service, role management and switchover, application service of standby library, protection mode of main library,DG the management.

Second, Redo Transfer Service

2.1 re-do transfer service introduction

Redo transport service coordinates the redo data transfer process from the main library to the standby repository. At the same time, the LGWR orl lns process will SGA "The buffered redo log is transferred to the repository; rfs receive lns lgwr write to the alternate redo log ( srl ).

Note:

1.ORL:Online Redo Logfile: Connection redo log file;

2.LNS:Log Network Server: responsible for transferring the redo buffer data from the main library SGA to the standby repository;

3.RFS:Remote File Server: responsible for receiving Data transmitted by LNS.

4.SRL:Standby Redo Logfile: Alternate redo log file.

2.2 Redo Transport Service Architecture

650) this.width=650; "Src=" Http://s4.51cto.com/wyfs02/M00/82/E5/wKiom1djvqbRlqMXAACDPJ_kRWs319.png-wh_500x0-wm_3 -wmp_4-s_4284624974.png "title=" redo transfer service. png "alt=" wkiom1djvqbrlqmxaacdpj_krws319.png-wh_50 "/>

Figure 2.1 Redo Transfer Service

DG supports two methods of transport for the LNS process: Synchronization method (sync), async method (async).

Synchronous Transfer to wait until LNS confirms that the redo data required for transaction recovery has been written to the disk on the standby site, the lgwr Approval commit operation is allowed to succeed. This approach protects all data from loss, but it can affect the performance of the main library.

Asynchronous Transfer , thelgwr does not have to wait for A confirmation message from the LNS, with little impact on the performance of the main library. In asynchronous transmissions, you can also enable the ability to redo data compression.

2.3 automatic Interval processing

650) this.width=650; "Src=" Http://s2.51cto.com/wyfs02/M00/82/E4/wKioL1djv8yjPwmjAAEAoJIk0r8820.png-wh_500x0-wm_3 -wmp_4-s_3188666658.png "title=" automatic interval processing. png "alt=" wkiol1djv8yjpwmjaaeaojik0r8820.png-wh_50 "/>

Figure 2.2 automatic interval processing

Log interval occurs when the LNS process stops transferring the redo data to the repository while the main library continues to commit the transaction. During the outage, a special ARCH process continuous ping is used to determine its status on the main library. When the communication with the standby is restored,theARCH ping process queries the standby control file through the RFS process to determine the last complete log file that the repository receives from the main library. Determine which log files are required to resynchronize the repository and transfer the corresponding log files through the other ARCH processes.

When the next log switch is performed, theLNS tries to connect to the repository and begins to transfer the current redo data after success, while ARCH handles the interval in the background. Application archive Log on the standby library after the speed of the redo record, the application process no longer reads the archive log, changing to read the current SRL(assuming "Live App" is configured).

Third, application services

Data Guard provides two different ways to apply the redo data on the standby database, respectively, as physical standby and logical standby.

Physical standby is the exact physical copy of the main library block by chunk, and logical fallback uses the logical process to apply the change orchestration to the standby database.

Note: Currently only used physical backup, logical backup did not go to contact.

Iv. Mode of protection

There are three types of DG protection modes: Maximum performance mode (default), maximum available mode, maximum protection mode.

Highest performance mode: Pay attention to the performance of the main library, despise data protection, and use ASYNC to redo the transmission.

Highest available mode: Most emphasis on usability, followed by zero data loss protection, using SYNC redo transfer. The main library failed to receive a confirmation message from the standby , the LNS connection is disconnected , and the transaction continues to be committed on the main library.

Maximum protection mode: Put data protection first and use SYNC to redo the transfer. If the main library does not receive a confirmation message for the available repositories, the main library stops.

V. Role Management and role transformation

There are two roles in DG, the main library and the repository, and there are two kinds of switchover and failover between roles .

Vi. Management methods of DG

There are three ways to manage DG, namely sql*plus,DG Broker and OEM.

Resources:

"Oracle Data Guard 11g Full Reference manual" chapter I.

[Reading notes] ORACLE11GR2 Dataguard Architecture