What is Recovery Point Objective (RPO)
- Tolerance for data loss
- How frequently should backups be taken ?
What is Recovery Time objective (RTO)
- Tolerance for downtime
- Downtime = Problem Identification + recovery planning + systems recovery
- Tiered RTO per level of granularity ( database, tablespace, table, row )
- Determine backup retention policy for on-site, off-site, and long-term backups.
- Retention policy can be configured based upon recovery window or Redundancy.
What is the criteria to assess the recovery requirements properly?
- You should first determine how critical a database is to the business when lost or unavailable?
- How much will the company loose per hour of downtime ?
- How much will be lost in productivity and labor costs if the database is down ?
How to classify databases according to criticality of availability (RTO)
- A large data warehouse can tolerate 12 hours worth of lost data, because batch loads can be re-run.
- An OLTP system cannot tolerate a downtime of even few minutes, as it results in the loss of $100,000 per hour of revenue. You will need to consider Data Guard here.
How to classify databases based upon how much data loss is acceptable (RPO)
- Are there databases that require more frequent backups ?
- Is there an adequate disk space available for archived logs ?
How to classify databases based on RTO?
- Is it minutes or hours ?
- Do you have a combined disk and tape backup strategy ? Based on this assess the retention period for backups.
No comments:
Post a Comment