Introduction to Oracle elastic uard

Introduction to Oracle elastic uard

 I. Basic principles of DataGuard

When a transaction changes the data in the production database, the Oracle database records the change in an online redo log file. You can configure the log Writing Process in the logging uard. In addition to logging logs to local online log files and archived log files, you can also use the network, send log information to a remote standby database server. This backup log file can be written in real time and synchronized to achieve zero data loss (maximum protection in the maximum protection mode); it can also be asynchronous, to reduce the pressure on network bandwidth (maximum performance in the maximum performance mode); or asynchronous and synchronous modes that can automatically switch (maximum availability in the maximum available mode ). After the backup database receives the log information, Data Guard can automatically use the log information to synchronize the Data with the master database in real time. When the primary database is active, the Standby database can perform the recovery operation. If the primary database fails, the Standby database can be activated and take over the production database.

 

Ii. Features of the three modes

Protection Mode	Data loss risks in the event of a disaster	Redo Transmission Mechanism	Whether standby redo log is required	Disk write
Maximum protection	Zero data loss	LGWRSYNC	YES	Affrem
Highest availability	Zero data loss	LGWRSYNC	YES	Affrem
Highest performance	Minimum Data loss-usually several seconds	Lgwr async or ARCH	Yes, but recommended	Affrem or noaffrem

Affrem: indicates that the redo log on the primary database is valid only when it is written to the standby log of the slave database.

 

NMaximum protection mode

The maximum protection mode provides the highest level of data protection for the primary database, thus ensuring a comprehensive disaster recovery solution for zero data loss. When running in the maximum protection mode, redo records are synchronized from the primary database to the standby database by the log writer (LGWR) process, transactions are committed on the primary database until transaction data is confirmed to be available on at least one backup server disk. We strongly recommend that you configure at least two slave databases in this mode. When the standby database is unavailable, the processing on the primary database will stop. This ensures that transactions are not lost when the primary database loses contact with all its standby databases.

Due to the synchronization feature of redo transmission, this maximum protection mode may potentially affect the primary database response time. You can minimize this impact by configuring a low-latency network and assigning it enough bandwidth to handle peak transaction loads. Enterprises that need this maximum protection model include stock exchanges, currency exchanges, and financial institutions.

 

NHighest availability mode

The highest availability mode is second only to the highest level of primary database data availability. Like in the maximum protection mode, redo data is synchronized from the primary database to the standby database by LGWR until transaction data is confirmed to be available on the disk of the backup server. However, in this mode (different from the maximum protection mode), if the last standby database involved becomes unavailable-for example, due to network connection problems, processing continues on the primary database, similar to semi-synchronous replication in the MySQL-5.5 ). Compared with the primary database, the Standby database may temporarily lag behind it. However, when it becomes available again, the Standby database will automatically synchronize the archived logs accumulated on the primary database without losing data.

Due to synchronous redo transmission, this protection mode may potentially affect response time and throughput. You can minimize this impact by configuring a low-latency network and assigning it enough bandwidth to handle peak transaction loads.

The highest availability mode is applicable to enterprises that want to ensure zero data loss protection, but do not want production databases to be affected by network/Backup Server faults. If another fault subsequently affects the production database and the initial network/backup server fault is resolved, these enterprises will accept the possibility of data loss.

 

NHighest performance mode

The highest performance mode is the default protection mode. Compared with the highest availability mode, it provides slightly less primary database data protection, but provides higher performance. In this mode, when the primary database processes transactions, the redo data is asynchronously transmitted from the LGWR process to the standby database. In addition, you can configure the archive process (ARCH) on the primary database to transmit redo data in this mode. Write operations on the primary database are completed in any case, and the submission operations of the primary database are not waiting for the standby database to confirm receipt (similar to asynchronous replication in MySQL ). If any backup target database becomes unavailable, the processing will continue on the primary database, which has a slight or no impact on the performance.

If the primary database fails, redo data that has not been sent to the standby database will be lost. However, if the network has enough throughput to keep up with the redo traffic peak and the LGWR process is used to transmit the redo traffic to the backup server, the lost transactions will be very small or zero.

 

Iii. Features of Oracle Multiple uard protection Modes

NMaximum protection mode

1). This mode provides the highest level of data protection

2). After redo logs are at least one physical slave database, the transactions in the master database can be committed.

3). When the primary database cannot find a suitable slave database to write data to, the primary database will be automatically disabled to prevent unprotected data

4). Advantage: This mode ensures no data loss from the slave database.

5 ). disadvantages: the automatic shutdown of the master database will affect the availability of the master database. At the same time, the master database must be restored before submission. The network and other objective conditions are extremely demanding, the performance of the master database is greatly affected.

 

NMaximum availability mode

1). This mode provides data protection capabilities second only to the "maximum protection mode ".

2). After redo logs are at least one physical slave database, the transactions in the master database can be committed.

3). When the primary database cannot find a suitable slave database to write data to, the primary database will not be closed, but will be temporarily downgraded to the "maximum performance mode" mode until the problem is solved.

4). Advantage: This mode ensures no data loss from the slave database without any problems. It is a compromise method.

5). Disadvantages: The disadvantage during normal operation is that the performance of the master database is affected by many factors.

 

NMaximum Performance Mode

1). the default mode provides the highest availability of the primary database.

2) ensure that the operations of the master database are not affected by the slave database during the operation of the master database. The master database transactions are committed normally and the operations of the master database are not affected due to any problems of the slave database.

3). Advantages: This avoids the impact of slave databases on the performance and availability of the primary database.

4). disadvantage: If the recovered data related to the transaction committed by the master database is not sent to the slave database, the transaction data will be lost, and no data loss can be guaranteed.