What is a reasonable uptime for your website?

All system administrators and blog writers dream about 100% uptime. However, is this realistic?
In this post, we bring an understanding of uptime SLAs and how to find the best for your scenario, whether if you have a business, a personal website or an app. We also present 2 possible solutions to decrease downtime and reach higher results.
What is a reasonable uptime?
Uptime is pretty much a relation of how much your provider invests in load balancing, high availability, good hardware and network, etc. In short, no matter how much is invested, 100% is never guaranteed. Even big companies face downtimes.
Understanding a reasonable uptime for each case depends on the cost of infrastructure and the much intended to avoid damage to your brand. Let’s first understand the most frequently offered SLAs in minutes/hours:

Thus, 99% of uptime means that your website might be down for 7 long hours in a month, or 87 hours in a year. It can be a fine for a personal blog, but not for e-commerce business. On the other hand, 99.99% of uptime means a possible downtime of 4 minutes in a month or 52 minutes in a year. This is much better, of course, but the costs must be considered.
Calculations
The best way to determine a proper uptime SLA is to produce a calculation of the cost you might have due to the unavailability of your website or server. Besides other costs, you must consider the loss of income (if clients cannot purchase on your website) and the loss of brand trust. Create an equation of that costs per time interval, and you will figure out easily if it’s worth it spending more on uptime guarantees.
Almost like a standard in the market, a 99.99% uptime guarantee is very common for a good price. However, if your brand cannot stand for 52min down within a year, you probably should invest in a better option.
And what about the 100% uptime promises?
There is simply no 100% guarantee. If your provider promises you that SLA, it’s not being much honest with you and you should question its credibility. In the long term, there’s no zero downtime. You can always expect a failure that will eventually affect your website.
If you are offered 100% uptime, it’s not much honest.
Even though companies like AWS and GoogleCloud invest millions of dollars in infrastructure, they don’t guarantee zero downtime. The following link presents outages faced by these big companies in 2017 — Tech Giant Outage.
2 steps to decrease downtime
But there are some ways to decrease the possibility of downtime of a server or website. You can eventually have multiple instances in different locations, with load balance, DNS tricks, etc. However, this solution is not simple, neither cheap.
Bellow, we present the two cheap non-intrusive steps that will help you mitigate the downtime of your website:
- Configure a CDN service: CDN is a very simple and useful solution that stands for Content Delivery Network. One of its main functions is to produce a distributed cache of your content. Even if your server is down, the CDN provider will deliver your website to the user. Some of the most popular CDN providers are CloudFlare and Amazon CloudFront. There are some limitations with dynamic content, but it works really fine with websites and blogs. You don’t need to touch your server; you only need to reconfigure your DNS to target the CDN provider.
- Configure a Web Monitoring Tool: the CDN cache is time-limited, and you cannot rely on it forever, in case of failure. You must take an action fastly to recover your servers, otherwise, the failure might be visible to users. To figure out any outage is happening, you must have a web monitoring tool to alert you as soon as a failure occurs. Web monitoring tools are non-intrusive as well, thus you can configure them without touching your server.
For that, you can count on Wideping to help you figure out when your website or server goes down. Sign up for a free account and start monitoring now.
