EXPERT RESPONSE
I'm not 100% sure of RTO, but I've used it to mean Return to Operation. In other words, this is the amount of time it takes to return the system to normal operation. The RTO time has a significant impact no your total up time. For instance, if you are required to have a 99.99% uptime, then you can only have 1 hour of downtime per week. For a one week period, your RTO is 1 hour otherwise you cannot meet your uptime level of service. When designing your backup and recovery strategy, you need to ensure you can meet your service level agreement, which means meeting a defined RTO period.
The MTBF acronym stands for Mean Time Between Failure. The MTBF defines the average amount of time that passes between hardware component failures. Normally, the DBA does not spend a large amount of time factoring in the hardware component's MTBF into their backup and recovery strategies.
For many, the MTTR acronym stands for Mean Time To Repair. Mean Time To Repair is normally thought of as a measure for the hardware components. Some piece of hardware fails and it takes a certain amount of time to repair…the MTTR.
For Oracle databases, the MTTR acronym stands for Mean Time To Recover. When an Oracle database abnormally terminates, the database needs to apply recovery before the instance can open. For a database that has a high rate of transactions, the time to apply recovery can be quite lengthy. This recovery time part of the RTO discussed above. In order to ensure the MTTR is a short period of time, Oracle provides the FAST_START_MTTR_TARGET parameter. When setting this parameter, Oracle will ensure that no more than FAST_START_MTTR_TARGET seconds are required to apply recovery. But setting this parameter too low can create performance issues with the database by creating too-frequent checkpoints.
|
