Understanding of HA ( High Availability ) of non functional requirement is not easy, it’s art with science. Being an architect, it is very important to understand cost vs acceptable downtime, i mean to say how much cost you are going to invest of servers and your acceptance downtime before committing to the stakeholder of your application.
High availability requires less human intervention to restore operation in complex systems, the reason for this being that the most common cause for outages is human error.
Availability Percentage and Downtime in a Year
Availability is usually expressed as a percentage of uptime in a given year. On high level following are the facts which
|Availability %||downtime in a year|
|90% ("one nine")||36.5 days|
|99% (two nines)||3.65 days|
|99.9% (three nines)||~ 9 hours (8.76 hours)|
|99.99% (four nines)||52.56 minutes|
|99.999% (five nines)||5.26 minutes|
|99.9999% (six nines)||31.5 seconds|
|99.99999% (seven nines)||3.15 seconds|
|99.999999% (eight nines)||315.569 milliseconds|
|99.9999999% (nine nines)||31.5569 milliseconds|
|~99.9998 %||1 min|
|~99.9886 %||1 hour|
|~99.7260 %||1 day|
if you go through the above tabular calculation data you can understand how much availabity you are required in your application and how much redundancy and cost have to pay for this Non-Functional requirement.
In general, the number of nines is not often used by a network engineer when modeling and measuring availability because it is hard to apply in formula.
1. In a year, what is the maximum number of days of downtime you can have and still claim >99% uptime?
a. less than 1 days
b. 2 days
c. 3 days
d. 4 days
e. 5 days
Answer: 3 days.