Downtime in the Cloud: How to Protect Your Business

Cloud providers are pretty aggressive when it comes to their SLA’s (Service Level Agreements) and promises of scalability and, more importantly, stability. When such promises are given, it is easy for online businesses to fall into a false sense of security. You assume that with such promises in place you shouldn’t have any problems with service availability. The reality of the matter, however, is that downtime plagues even those in the cloud. Here are a few things to keep in mind and ways to protect your online business from downtime in the cloud.

Remember the Cloud is Just a Part of the Picture

Downtime is defined as any service interruption to your customers. Your cloud infrastructure can be up and running perfectly, but if something goes awry, such as a bug in your script or a third-party app affecting the functionality of your entire site, you face downtime issues. Therefore, it is important to remember that your cloud provider’s downtime is only a small piece of the entire downtime puzzle, and you need to take measures to ensure that your online business is protected on all fronts – not just the fronts presented by your cloud provider.

Understand that SLA’s Are Not Always Met

The SLA that your cloud provider offers may promise uptime availability of 99.9 percent, which equates to approximately 43 minutes of downtime per month. However, you must ask yourself, does an SLA with a promise of 99.9 percent uptime really mean that your site will be up 99.9 percent of the time? Not necessarily. Things still can, and do, go wrong. The December 2012 AWS outage of over 20 hours is proof of this fact. When things do go wrong, all you can do is apply for a service credit that is quite meager compensation when compared to the drastic losses your business would face from such an outage. While the SLA provided by your cloud service provider may offer you a sense of security and can be used as a guideline for what type of uptime you can expect, it is important to realize that it isn’t set in stone nor does it offer any real protection for your business.

Plan for Failure

It may sound counter-intuitive, but planning for failure and ensuring that you are working with a cloud provider that has built their services on a dynamic infrastructure that can withstand failure is an important step for surviving cloud downtime. By taking failure into account when choosing a cloud service provider, you can ensure that the provider you choose is one that is less likely to experience extended downtime issues.

The data centers of large public cloud providers are usually built across zones that are known as “availability zones,” which are various locations engineered to be insulated from failures that occur across one another. The concept behind this is that if one zone fails, users can be redirected to another zone in real-time. If you want to protect yourself from downtime in the cloud, make sure that you utilize a cloud provider that offers such an infrastructure. While you may experience some performance issues if your zone does go down, you will be able to avoid a complete downtime outage.

Utilize Multiple Providers

While utilizing a cloud provider that has various insulated zones will protect you from downtime in the cloud, it can’t entirely eliminate your downtime risk. Because of this, you may want to consider using multiple cloud providers. However, it is important to remember that this approach is not always foolproof since many cloud providers share common data center resources. Because of this, you need to check with the individual providers you are considering as a secondary provider to ensure that they aren’t sharing resources with your primary cloud provider. Otherwise this redundancy protection isn’t really offering you any protection at all.

Implement a Monitoring Service

Even the best laid plans can’t always prevent downtime in the cloud. When downtime does occur, you need to know about it the moment it happens. This means putting monitoring services into place. You need to utilize a service that will offer you 24/7 monitoring and will notify you with alerts sent via SMS text message, phone calls, and emails the moment you do experience an outage. This will allow you to put backup plans (such as your secondary cloud provider) into place to prevent unnecessary extended downtime periods.