I'm pretty sure that at some point in all of our careers, we've experienced the hassle of a system outage. The frustration of being able to do nothing but watch as a technician is called out, working his way around each machine, running diagnosis' and upgrades.
Hours later he emerges from the 'server room' with a hefty bill and the instruction to wait at least 2 hours for the system to reboot.
Days like these can be crippling; not just for morale and employee momentum but productivity and revenue. And depending on the size of your business, the financial impact can reach big numbers. But for business of all sizes, the impact can go on to harm your productivity for weeks after.
Uptime = revenue
These days, technology is the ultimate enabler, and as a consequence, it’s relied on by businesses like never before.
So, if IT systems are unavailable, it can have a detrimental effect on productivity. According to City Talk, more than two thirds of UK businesses have experienced internet connection failures, costing the economy £12.3 billion in lost productivity and extra overtime.
With inevitable outages, upgrades, and system crashes, traditional local-hosting models come with an undesirable amount of downtime. The consequences of which extend further than just the impact on profitability. User confidence and loyalty, company reputation, employee productivity, and company morale all take a hit. Not to mention the labour expense, system restoration and other associated costs that come when your server takes a turn for the worse.
The uptime of the cloud will never reach 100%; even cloud services are susceptible to network and other outages. And, according to Werner Vogels, CTO of Amazon.com “Everything fails, all the time.” But the best cloud providers put significant measures in place to achieve the coveted 99.999% uptime (often known as the 'five nines').
That percentage of downtime equates to around 5 minutes per year - roughly 26 seconds a week. Historically, the average traditional solution achieved on average between 99% and 99.5% uptime, which doesn't sound too bad. But when you realise that that equates to between 9 hours and 3.5 days downtime per year you begin to understand why providers are continually striving to improve their uptime stats.
Mitigation against downtime
Outages are inevitable - even with cloud computing. But what it comes down to is how your provider guards against risks and how quickly they react when one occurs, recovering the issue.
The best cloud providers put measures in place to mitigate against downtime disruption. You should look for the following features in their offering to understand how vigilant they are about maintaining high availability:
Redundancy refers to the duplication of critical components and more than one virtual instance for your apps and data. If your primary instance goes offline there is a reliable second system which kicks in.
It goes without saying that security should be a priority in any case but particularly when guarding against cyber hacking and attacks.
Attacks of this nature can deem a system unavailable for an unspecified amount of time, but they be prevented by having strong security in the form of multiple firewalls and antivirus programs.
Tier IV data centres
Providing a multi-layered approach to security begins with the physical security of your data. Tier IV data centres are especially rigourous - enough to withstand a Category 5 hurricane. At the other end of the scale, a Tier I data centre has no built in redundancy to prevent disruptions, which have an enormous impact on availability.
Efficiently distributing workloads across multiple computing resources is what is known as load balancing. This way the system guards against overloading a single resource, which could result in poor performance for users or even system failure.
Security, natural disasters, and inevitable power outages are all to blame for the loss of valuable time. And unfortuantely, time is a commodity that can't be recovered. That's why mitigating against the likelihood of downtime is the first step to uptime success. Armed with the above knowledge you’ll hopefully feel more confident in which questions to ask potential vendors, and which features to prioritise before you make your decision.
Multiple deployment models of the cloud have been designed to suit every business model. You can start your cloud journey with three simple steps. Read our Cloud Journey ebook find out more.