How to Calculate Uptime

Everyone who has ever worked with web resources as the promotion specialist or full-fledged webmaster surely knows what uptime is. If you’d use a simple definition, uptime is the server availability metric that's counted with the basic system availability formula - overall lifetime minus unavailability time, but shown in percentage instead of actual minutes and hours. Therefore, it could become lower from any issue, even the maintenance routine, that affects the network or the server itself. And, obviously, the higher it is - the better.

This should be known by anyone who wants to successfully promote their website to the top of operation stability, round-the-clock availability for users and, following all that, benefits and profits for business owners. Because you won’t be able to achieve any of the aforementioned things without having at least 99.9 uptime in hours and striving to reach 99.99 uptime in hours or even 99.999%. This sounds rather harsh, but maintaining the highest possible uptime is one of the cornerstones of creating well-made and successful websites. Moreover, all search engines and ads platforms are using the same metric as a basis for appraising the reliability of the resource.

Without taking uptime into account, it would be impossible to maintain your website growth and service stability, keep and attract an audience and make it profitable. In other words, availability and uptime are highly vital for every web resource. Therefore, it’s a must for web resource owners to properly understand how calculating availability works and how uptime monitoring works.

Five nines meaning

No matter how hard you’ll be monitoring and maintaining the website’s uptime, the most impossible goal you can have is reaching 100% uptime, and there’s the reason for it. This is impossible because your server needs maintenance and rebooting, Not to mention various emergencies like power outages and network failures - all of them will affect the uptime and it will go down. But you can actually strive to reach 5 9s uptime or ‘almost 100% availability’ metric.

The 99.999 uptime means literally what it says - it’s when your web resource is up and running under optimal conditions for 99,99% of its overall lifetime. This number only looks unreachable, but it’s actually pretty realistic if you have a reliable server or hosting, and it could work without any issues. Of course, you’ll need a bit of luck to avoid various unexpected disasters that you can’t do anything about. Also, there are a bit more realistic numbers if you want to set an SLA goal.

Basically, your uptime report shows 3 9s meaning that your website can actually be considered highly accessible, but you need to raise the uptime percentage by all means. The higher it is - the better experience users will have from visiting your web resource. And that’s when your only solution is to get a good uptime software for calculating it and monitoring it 24/7.

Uptime calculation

As it was mentioned earlier, the basic uptime calculation formula is lifetime minus website downtime shown in percentage:

• (Overall Lifetime - Overall Downtime) x 100% = Overall Uptime

But this equation is only valid when you’re calculating server uptime. For SLA metrics this won’t be needed - you need to count only the time your resource lives, and it could be hosted on the long-living hardware with years of prior lifetime.

How to calculate downtime

To calculate downtime, you need to count every situation whenever your resource was inaccessible from the outside and unreachable for users. Up to seconds. And it’s best to be a number in seconds, because minutes, not to mention hours, of downtime could drop uptime percentage to its lowest. Obviously, it would be hard to recover from such situations.

Calculate uptime

As it was mentioned, the basic uptime formula is not really valid for web resources except servers. Therefore, usually website uptime calculation is going in accordance with the monitoring period, because you can’t possibly know for how long the website existed before you started monitoring it. 

How to measure uptime

Taking into account that in most cases, uptime calculation includes only time when you’re monitoring it. Therefore, you use the following uptime formula:

• (Monitored Time - Monitored Downtime) x 100% = Uptime

In other words, commonly you calculate uptime with time from the beginning of the monitoring minus downtime from the beginning of the monitoring. All shown in percentage.

Availability metrics

The only satisfactory uptime score that you should strive to achieve is, as it was mentioned earlier, is Four Nines or higher. Because it’s literally nearing perfection, and it’s a vital sla uptime goal. Although it could be pretty hard to achieve it, there are ways to do so:

• Find a hosting with good reliability. One of the metrics of the reliable hosting service is its server uptime. Obviously, it should have a percent close to 99.99 or even above. And it’s vital to check service availability by yourself, without blindly believing in hosting advertisements.

• Find a good server uptime monitoring service. It should automatically monitor uptime, make regular reports and notify you or your webmaster about problems threatening uptime stability. For instance, Host-Tracker is one of the best services in the niche, and it can do a lot more than checking uptime.

• Minimize web resource and server update, reboot and maintenance time. As it was mentioned, every second counts. Update shutdown, or even a fast reboot, takes from availability time, and it will affect uptime. So it needs to be done as fast as possible.

• Use backup servers or cloud platforms. A good way to avoid lowering web resource uptime availability is to run it on a separate server or in the cloud while conducting maintenance on the main server. It’ll lower the resource’s downtime to a minimum and give you more time for maintenance without interrupting uptime.

Taking all the above into account, you can avoid lowering your website uptime by using reliable hosting, conducting constant monitoring, minimizing maintenance and using a backup server or cloud platform if possible.

FAQ

Which formula is used to determine a cloud provider's availability percentage?

Cloud providers, taking into account that cloud is a server cluster that should work 24/7, must have 99.99% uptime and availability or higher, using the following formula: (Guaranteed Uptime - Downtime) / Guaranteed Uptime.

What measures the average amount of time between failures for a particular system?

Average time between incidents or failures (MTBF - Mean Time Between Failures) is measured with the following equation: Overall Operational Time / # Of Failures.

If a network has the five nines of availability, how much downtime does it experience per year?

For every web resource or network, Five Nines or 99.999 uptime percentage, means that it has less than 360 seconds (86400 seconds/day x 365 days x 99.999%) of average annual downtime. 

A highly available server is available what percentage of the time?

‘Highly Available’ is the definition that’s applied only for resources, servers included, that achieved Five Nines, or 99.999 uptime percentage.

What is the downtime per day for a system showing 99.9% availability?

Three Nines. or 99.9 uptime percentage, means that the resource has less than 131 minutes (1440 minutes/day x 365 days x 99.9%) of average annual downtime.