Just as people used to keep money under their mattress, untrusting of banks, some organizations today keep their data on disks, wary of the cloud. But those organizations are few and far between, and cloud adoption rates continue to soar across every industry, with experts predicting an 85% adoption rate in 2019. Despite this, the question of reliability persists. Just as people used to wonder, “will my money be available at a bank when I need it?” today the question is, “will my data be available in the cloud when I need it?”.
The Azure Incident
In September 2018, one of Microsoft Azure’s data centers in San Antonio, Texas, went dark due to a “severe weather event, including lighting strikes.” This incident left Azure customers located in the South-Central US unable to connect to their resources for hours during the downtime; some customers not even in the region experienced trouble with Azure Active Directory, Visual Studio Team Services, and Azure Resource Manager. It’s large scale problems like this that break the trust people have in the cloud, causing them to rethink their decision. These incidents also reaffirm for those who’ve not yet migrated that they’ve made the right choice. But do isolated incidents like this really mean the cloud can’t be trusted?
What’s in a Nine?
When shopping around for a cloud provider, the most important question you can ask to get a sense of their reliability revolves around uptime. As its name implies, uptime refers to how fast a data center can get your business, your applications, and your website up and running in the event of incident or interruption.
The industry standard for uptime is measured in percentages based on how many nines are included. A starting point for some cloud providers may be 99% (“two nines”), which is the equivalent of about 3 days of downtime per year. This will certainly make for an inexpensive deal, but may be too much downtime for some organizations to even consider. 99.9% (“three nines”) reduces downtime to about eight hours per year, which may be adequate for some businesses that can handle a brief loss in productivity throughout the year, but for healthcare providers and government agencies that need immediate access to data, it is still considered inadequate. 99.99% (“four nines”) further reduces downtime to under an hour per year—just a few seconds per day—which is where most organizations find their comfort level.
A Recipe for Disaster Recovery
A disaster can strike when we least expect it (just ask Azure’s Texas data center). Take a look at these eye-opening US statistics:
The average cost of a cyberattack for enterprises in 2017 was $1.3 million ($117,000 for small and medium-sized businesses)
Electrical power outages, surges, and spikes are estimated to cost more than $150 billion in annual damages to the United States economy
2017 was the most expensive hurricane season in U.S. history, costing roughly $200 billion in damages
That’s why preparation is so important to reliability. Incidents such as these need only be minor setbacks if a cloud provider puts the following safety protocols in place:
Uninterrupted Power Supply (UPS)
Computer Room Air Conditioning (CRAC)
Encryption for data-at-rest and data-in-flight
Safe proximity from flood zones
Hurricane structure rated facilities
24/7 surveillance and physical security (HID card, PIN, biometric access)
Do Your Due Diligence
You should always do some research and investigation into a potential provider’s track record, as they may not always be forthcoming with information. What is the longest they’ve gone without downtime? Have there been any disasters in the past that damaged their reputation or resulted in a loss of customers? Regardless of your findings, the provider should guarantee their services by providing a Service Level Agreement (SLA) which details their responsibilities, and the penalties for failing to live up to them. Be sure that you are comfortable with the SLA offering from your cloud provider—and always use your negotiation skills.
When it comes to critical data, no one likes surprises. And while there are bound to be occasional hiccups with any provider, through proper research and planning, you can find a provider that will mitigate downtime and protect against disaster. And, considering all the benefits of cloud computing, the odd malfunction can be a relatively small price to pay.
DSM is Florida’s predictable cloud provider. With DSM, you know what to expect, with no surprises—which is what everyone wants. You can count on reliability when it comes to service, security, support, compliance, contracts, and pricing. Learn more about finding peace of mind with a predictable cloud here, or contact one of our experts at 877-376-6381.