Why the AWS outage has some questioning the internet’s codependence on hyperscalers

Pills that are hard to swallow: Not everyone will like you, life isn’t always fair, and the internet may rely too much on a handful of tech providers that can crash everything if they go down.

This week’s outage at Amazon Web Services (AWS) has some experts questioning the sustainability of the internet’s dependence on a small number of cloud giants.

The TL;DR of what happened. On Oct. 20, AWS seemingly broke the internet when it began to experience “increased error rates and latencies for multiple AWS services in the US-EAST-1 Region” in the early hours of the morning. The cloud giant said the outage was caused by a domain name system-related issue.

Because AWS is practically the backbone of the internet, millions of users were caught in the crossfire of the outage. Apps and popular websites like Snapchat, Venmo, and Roblox felt the impact, with each experiencing a spike in Downdetector reports.

“Yesterday, I was at a restaurant and I couldn’t pay because their system was down,” Catchpoint Systems CEO Mehdi Daoudi told IT Brew of his own experience with the outage, adding that the staff didn’t know how to charge his card manually, either.

The outage had more detrimental effects, as well. Critical infrastructure got whiplash from the cloud giant’s downtime as banks and airlines experienced disruptions on their apps.

The epiphany. By Monday evening, AWS said all of its services had returned to “normal operations.” However, the incident had already made one thing clear, according to Samir Elabed, a cybersecurity engineer: “It shows us that we rely so much on one provider, which is in terms of business planning, I would say, no, we can’t do that anymore.” AWS had a global market share of 30% in Q2, as per Synergy Research Group data.

Patrick Neighorn, an AWS spokesperson, told IT Brew in a statement that AWS has “maintained a track record for reliability that surpasses other major cloud providers,” but did not comment on concerns related to the outage.

And AWS isn’t alone in shouldering that responsibility for uptime: Microsoft Azure has 20% of the cloud infrastructure services market, followed by Google Cloud at 13%.

So, now what? While it may feel as if the whole internet has put its eggs in the Big Three’s basket, several professionals say all is not lost. Daoudi suggested the AWS outage is a wake-up call that should inspire businesses to consider hiring a chief resilience officer tasked with maintaining IT infrastructure resilience.

“The only solution is to start thinking about resiliency at the board level,” he said.

Elabed thinks companies should adopt a multi-cloud or multi-region cloud strategy.

“There is the saying, ‘AWS never goes down,’ but look what happened yesterday,” Elabed said, adding that a multi-cloud or multi-region strategy would give a business a backup plan if another outage were to occur.

Spencer Kimball, CEO of Cockroach Labs, shared similar thoughts, adding that each of the hyperscalers is a “technical monoculture.”

Kimball said businesses should “straddle” the different cloud providers. He also referenced the EU’s respective Digital Operational Resilience Act and Data Act, which aims to bolster the digital resilience of financial entities while ban egress fees, which can make it more expensive for a company to pursue a multi-cloud strategy.

“That’s a very interesting regulation,” Kimball said. “Expect to see more of those, 100%, and things are absolutely moving in that direction.”