October 12, 2018

Potential Perils of Depending on a Single Cloud Hyperscaler

The Risks of Depending on a Single Hyperscaler

As digitalization progresses, enterprises are increasingly relying on the cloud for simplicity and efficiency. According to Gartner, the worldwide public cloud services market is expected to grow 21.4% in 2018, bringing it to a total of $186.4 billion.

While the public cloud brings promises of numerous benefits, such as reduced operational costs, improved workforce productivity, operational resilience, and business agility, it has to be used strategically. Migrating all of your workloads into a single hyperscaler may not be the best move for getting the most out of the cloud. It’s a great option to offload hassle and costs, but you need to choose the right clouds and redundancy options for your needs.

A few reasons why putting all of your eggs in one cloud can be risky:

  • Vendor lock-in
  • Risk of DDoS attacks
  • Lack of reliability
  • Potential for downtime (CIO from IDG)

Why Your Cloud Strategy Must Include Redundancy at Multiple Levels

Now, we’re not discouraging the use of hyperscalers. We’re just saying that having nimble options for redundancy alternatives in your data center is ideal.

Let’s review the recent Microsoft Azure outage that, due to a lack of availability zones, caused an automatic shutdown of hardware. After lightning during a storm caused a voltage swell in utility feeds that powered one of the Azure data centers in Texas, cooling systems were knocked out, and as air temperature rose, the hardware shutdown was triggered. Ultimately, the outage affected close to 40 Azure services in the South Central region, as well as multiple other locations. Unfortunately, those services either relied on the storage systems that failed or had dependencies with services that relied on them (Data Center Knowledge).

The above example took place before Azure introduced availability zones, and demonstrates the potential consequences of relying on a single cloud. It is key for businesses to make sure their cloud strategy includes redundancy at multiple levels. Using a single public cloud provider does not guarantee redundancy, as evidenced by the Azure outage.

Enabling a Cloud Redundancy with Data Center Cloud Connections

If you’re looking to get the most out of the cloud, broadening your portfolio may be a good tactic for putting your environment in a more resilient position. Relying on a public cloud connection through a reputable data center is one way to establish a single interconnection point for cost-effective multi-region and multi-cloud redundancy.

Cloud Direct Connect, for example, is a secure, fully private connection between customer infrastructure located in DataBank data centers and hyperscale cloud providers. DataBank is partnered with Megaport, a leading provider of SDN-based elastic interconnection solutions, to facilitate access from DataBank data center locations to a wide range of providers on the Megaport network. With Megaport-enabled Cloud Direct Connect, DataBank colocation or managed services customers can connect directly to a range of public clouds.

If you’re looking for a way to establish better redundancy, DataBank can help you strategize. Contact us online or call us at 1.800.840.7533 and speak to a cloud specialist.