Bandwidth qos of Oracle dataguard

[Former Member]

We are going to build DR system by using dataguard.

The bandwidth between Primary DC and DR DC is 100Mbps, it is enough for redo-log apply in normal sync mode, but when we initial the DR database, there are tones of redo logs and it will used up the bandwidth without network qos control.

I know the best practice is to set qos on network level between primary DB and standby DB, but I am curious whether we can set qos parameter on oracle level for dataguard redo log apply?

Thanks,

James

[Former Member · Accepted solution]

We do NOT want to send the log asap, instead, we need control DG to use part of the bandwidth.

http://www.oracle.com/technetwork/database/features/availability/maa-wp-10gr2-dataguardnetworkbestpr...

In page 8, TCP socket buffer settings will control how much network bandwidth can be used regardless of the bandwidth available in the network circuit.

If we lower the socks port number, then bandwidth utilization can be slow down? We will test later.

[Former Member]

Hello Lu,

My colleague has released 2 useful documents about this subject. You can have a look and find your answers and more in those documents:

I hope they help!

Regards,

Serhat

[stefan_koehler]

Hi James,

> but when we initial the DR database, there are tones of redo logs and it will used up the bandwidth without network qos control.

... and this is one reason why you should use "RMAN Incremental Backups to Roll Forward a Physical Standby Database": Using RMAN to Back Up and Restore Files

Regards

Stefan