By Kayla Matthews, tech journalist & blogger
Remember last February when it felt like the internet was broken? Sites including Netflix, Spotify, Pinterest, Buzzfeed and many others experienced downtime.
Why? Because Amazon’s Web Services — a cloud computing service — incurred an outage for about four hours. Considering that thousands of websites rely on Amazon’s services to function properly, a decent chunk of the web was affected. The S3 system for Web Services went down, which is used by over 160,000 sites according to SimilarTech.
This may come as a surprise to some, who associate Amazon with retail and online shopping as opposed to cloud computing and web hosting services. In reality, Amazon’s service is a lucrative opportunity for businesses worldwide, big and small. It gives them the option to deploy cloud computing and online services, without investing in their own server farms.
Amazon, Server Farms and Your Business
A company looking to offer online services to customers or wanting to store a great deal of data would also have to invest in the necessary hardware — a server farm. This is a series of computer systems designed to host web-based content, but it’s expensive. It takes a lot of power to keep a server farm up and running.
The cheaper option is often to host content through a third-party service like Amazon’s. This outage does offer some perspective, however. Namely, it reveals what can happen when a service you rely on goes down. At the very least, it will frustrate customers while wasting thousands — maybe even millions — of dollars in downtime.
If you’re planning to use a hosting service like Amazon’s, you could stand to learn a few things from this event.
- Always Have a Backup Plan
Amazon Web Services were only down for about 4 hours. Imagine if they had been down longer?
In many cases, web services and online content are all that companies are serving their customers. If they can no longer access these goods because of an outage, it can cause a lot of problems. For starters, it means revenue can slow or even stop completely — especially for online retailers.
That’s why it’s always necessary to have a backup plan. What do you do when the services you rely on are down? What can you do if they remain down for an extended period? There are temporary solutions for hosting content in the meantime, and it may be a good idea to subscribe to more than one hosting service. Of course, that would mean you’ll need to keep two — or more — up-to-date versions of your cloud content at all times, which can be a hassle to maintain.
Still, it’s tough to overlook the benefit of little to no downtime if and when something happens.
- Minimize Cloud Dependency
Okay, so, your business needs cloud services to function, but you don’t need to have absolutely everything hosted in the cloud. It is possible to have a small portion of your content hosted, while the rest is stored elsewhere.
The goal is to figure out what is mission critical. In the event of an outage, what systems and content must absolutely remain accessible? In the case of complete web platforms like Netflix, this is a tough egg to crack. But if you can scale so only a few of your functions or content are outsourced to a cloud platform, you may be in better shape when something like this happens.
Yes, prepare for when. Let’s be honest: It’s happened before and it just happened in February, and it will happen again.
- Be Prepared
To understand why risk-based thinking is a key component of some ISO standards in business, consider how it helps companies prepare for bad things while minimizing losses. Best practice is to prepare before an outage happens — when you can function without panicking.
The worst thing that can happen is you’re trying to figure out options to get your services or content back online temporarily, right in the middle of an outage. In advance of a problem, consider these questions:
- Do you have a monitoring system that will identify when your web services or content are down?
- Can you quickly identify whether it’s a problem with your hosting provider or on your end?
- What can you do during an outage to continue operating?
- What tools and services do you need to have prepared?
Make sure your backup system is in place and you have procedures outlined for your team so they know what to do during an outage. These are all important parts of disaster preparedness.
- Read and Understand Your SLA
Service Level Agreements, or SLAs for short, are like a contract between the business and cloud provider. It details the type of service you can expect, but it should also explain how you can be compensated when there is an outage.
Let’s say an outage happens right in the middle of one of your biggest sales ever, costing you millions of dollars. You’re going to need to recuperate some of those losses — or at the least enough to cover operating costs during the outage — because you weren’t bringing in revenue. But this is only possible if an issue goes against the agreement.
An SLA is laid out beforehand to direct responsibility to the appropriate parties. You don’t want to be stuck in the middle of an outage with your service provider telling you “sorry, you’re out of luck.” The problem is, it’s not the type of contract you have the authority to revise and edit. Once you sign on the dotted line, that’s it. So, make sure you read it carefully before diving into the deep end.