The Public Cloud: How IT Leadership Can Manage Resiliency, Outages

While it may be more reliable than your internal data center, putting your infrastructure in the cloud is also putting it out of your control. In the event of an outage, you are left fielding phone calls from outraged users instead of actually working on creating a fix to the problem.

Outages at cloud computing providers is a top concern among IT leaders according to analysis by NetworkComputing and InformationWeek. A majority (52%) rated outages at cloud computing providers as a top concern when it comes to how it impacts network operations.

But just how often do major outages really happen? Network Computing 's Salvatore Salamone recently broke that down, in an article looking at recent major outages at AWS, IBM Cloud, Google Cloud, Facebook, Microsoft Azure, and Oracle. It’s not so reassuring when a minor issue like a cut cable can turn into a big problem that impacts the operations and finances of multiple organizations.

"The numerous outages over the last year have caused wide-scale disruption to businesses worldwide. Most were caused by configuration changes done by the providers themselves, a handful were due to mother nature, and some were due to long-standing familiar issues like power outages."        

Everyone is more reliant on the cloud now. The big cloud providers AWS, Google Cloud Platform, and Microsoft Azure, are still growing at a rapid pace, even if that slowed down a little bit in Q2.

You may wonder if you can mitigate your organization's risk with an analysis of the cause of major outages. Network Computing took a closer look at the causes of disrupted network availability. Here's what the research says:

Cloud providers talk about how they are architecting data centers to prevent climate change. But are they building them to be resilient from the impact of climate change like extreme weather? A really interesting take on this emerging issue is provided here:

It is a real concern, according to recent research by Network Computing and InformationWeek

Whatever the cause, outages are inevitable. It's best to be ready for them when they do happen. InformationWeek contributing editor John Edwards provides some perspectives on what CIOs can do to prepare for outages here:

CIOs' growing sophistication concerning tradeoffs between costs, risks, and resiliency will impact choices they make. Here's an inside look at how they've responded to big outages in the past, and how their understanding and approach to outages has changed over time. Learn from their experience here:

For instance, it's a big deal at a retail company when their cloud provider experiences an 8-hour outage during the Christmas shopping season as happened in 2021.

“We definitely had a lot of customers who panicked. This was in the middle of Christmas season. There were retailers who couldn’t sell things. There were banks that couldn’t process transactions, largely because mobile was down.” -Brent Ellis, Forrester Research.        

CIOs expect resiliency and get a bit upset when they haven’t gotten what they expected, according to Omdia 's Roy Illsley

"There’s a balancing of expectations of what’s included in the cloud, what’s excluded, and where additional costs and risks come in." - Roy Issley, Omdia.        

While some tactics have been rejected by IT leaders due to their complexity or high cost, here's a breakdown of some emerging technology that could help IT leaders prepare to mitigate the damage from cloud outages, as described by InformationWeek contributing editor Pam Baker in this piece:

You may wonder, if your first-choice cloud provider experiences a failure, why can't you just set up to fail over to your second-choice cloud provider? Forrester 's Brent Ellis provides the analysis of the factors organizations should weigh as they consider that option.

If there's a sustained cloud outage, can you recover your losses? How do you figure out what it costs you per minute of downtime? What do your cloud contracts say about it?

Not many organizations are big enough to negotiate contracts that provide for compensation, according to Joseph Williams , partner of cybersecurity strategy at Infosys Consulting.

Another way to get compensated is with insurance coverage. But the price of coverage has increased due to the increasing complex loss environment says Michael Phillips, Chief Claims Officer at Resilience

What about a lawsuit? Can you sue? Should you sue?

If you are a network manager, a lot of the planning and remediation may fall to you. Better observability tools can help but but provider misconfigurations and DNS infrastructure issues aren't anything you can control, as Network Computing reports here.

The catastrophic consequences of cloud outages has not gone unnoticed by governments. The US, UK, and EU are all weighing regs to classify cloud companies as"critical infrastructure" and require they meet resiliency standards. Here's a look at what's happening in this important policy arena.

As long has there has been cloud computing, there have been outages. InformationWeek 's Editor-in-Chief Sara Peters takes us through the archives to look at ye olde outages of yore.

Meanwhile, cloud providers themselves must weigh market pressures, costs, and risks as they take measures to ensure their future resiliency. Here's a look at that aspect by InformationWeek 's senior writer Joao-Pierre Ruth .

Interested in more Big Picture overviews from the InformationWeek team? Be sure to subscribe our email newsletters.