You may hear the phase Five-9s when it comes to the availability of a database. This concept is linked to how long your database is up for or your KPI (Key Performance Indicator) for the product's uptime.
You may boast your database is up 99% of the time and think that sounds good, but if you do the maths over a year it doesn't look so great.
This Microsoft table looks at the uptime depending on how many nines you include?
Number of 9s | Availability Percentage | Total Annual Downtime |
|---|---|---|
2 | 99% | 3 days & 15 hours |
3 | 99.9% | 8 hours & 45 minutes |
4 | 99.99% | 52 minutes & 34 seconds |
5 | 99.999% | 5 minutes & 15 seconds |
So, as you can see, saying you are up 99% of the time means your database is unavailable for over three days a year. This is why so many companies are aiming for Five-9s.
Principles of Availability
When you are looking for high availability there are three areas to consider:
Single Points of Failure (SPoF)
Redundancy
Monitoring
Types of Availability
In general there are five different approaches to providing high availability:
Replication
Log Shipping
Mirroring
Clustering
Tools - these are applications that are written, sometimes for a 3rd party, to help keep your database functioning. For example, MS SQL DB has AlwaysON
Related Posts
