Previous

Content  

Next

1.1. Availability  

The first measure of "Quality of Service" is availability. Perhaps people in charge of the network, or your ISP, will tell you: "our service is one of the best all over the world, we are available 24 hours per day, 7 days per week, 52 weeks per year". If you can believe in this asseveration, the service of your provider is available 100%, this mean, at any moment any time you need it.
But, normally, things are not as perfect at all, and when you need more your connection, because you must send a very important e-mail, at that moment, exactly, your connection is out of service and you are out of business.
Looking things from your point of view it’s difficult to establish a measure for availability. Let us suppose that your provider is honest, they take their measures, and they can tell you that last year the service they provided was available 98% of the time. Making some calculation, you say:
24 hours/day x 365 days x 0.02 = 175.2 hours  

Okay. The service was not available 172.5 hours the last year. Let's now suppose that the service this year is going to be as good as it was the last year, this mean, 98% of availability.
But, what about if you need to send your very important e-mail just in one minute of those 175.2 hours that the service is out? Are you going to evaluate your ISP "Quality of Service" as they say, 98% of availability?
Ignoring the fact that having the connection, or not having it, is quite different depending on your urgency, we could take the approach of establishing that, from "your" point of view, availability could be measure as the relation between number of times that the service was available, divided by number of times you needed it.
This way, if the last year, for example, you needed the service 514 times and it was available 485 times, the real measure for availability, from your point of view, was just 94.35%, independently that the ISP has measured 98% using its special metric.
   

Previous

Content  

Next