Digital dependence: AWS outage exposes global cloud weakness

The world’s largest cloud computing platform, Amazon Web Services (AWS), has suffered a major outage that disrupted thousands of organisations, including banks, financial software platforms such as Xero, and social media platforms such as Snapchat.

The outage began around 6 pm AEDT on Monday and stemmed from a malfunction at one of AWS’s data centres in Northern Virginia, United States. Although AWS says it has fixed the underlying problem, some users continue to experience service disruptions.

The incident underscores the vulnerabilities of a world that depends so heavily on “the cloud.” Yet there are ways to reduce these risks.

Cloud computing refers to the on-demand delivery of IT resources — such as computing power, data storage, and applications — via the internet. In simple terms, it allows businesses to rent, rather than own, their IT infrastructure.

The model took off during the dot-com boom of the late 1990s, as digital companies began providing software online. As firms such as Amazon grew more capable of offering “software as a service,” they began renting out their virtual servers to others — for a price.

This created a powerful value proposition: instead of the heavy upfront costs of running data centres, companies could pay for IT resources as they used them, much like a utility bill. Today, more than 94% of all enterprises use some form of cloud-based service.

The global cloud market is dominated by three providers. AWS holds about 30% of the market, followed by Microsoft Azure with around 20% and Google Cloud Platform with roughly 13%.

Each has suffered significant outages in recent years. In 2024, a third-party software issue severely disrupted Microsoft Azure, causing operational chaos for businesses worldwide. Google Cloud Platform also experienced a major outage due to an internal misconfiguration.

This concentration of control in just a few major providers poses serious risks for businesses and users alike.

First, it creates a single point of failure. As the AWS incident shows, one technical glitch in a central system can cascade through the internet, paralysing key services in minutes.

Second, these providers often lock customers into their ecosystems. Migrating data between platforms is technically complex and prohibitively expensive, largely because of high “egress fees” charged for transferring large volumes of data. This vendor lock-in effectively traps companies with a single provider.

Third, the dominance of US-based cloud companies introduces geopolitical and regulatory risks. Data hosted on their servers falls under US jurisdiction, complicating compliance with other nations’ data sovereignty laws, such as Australia’s Privacy Act. It also gives these companies potential leverage over who can access or distribute information — a subtle but real form of control.

To reduce these risks, experts recommend adopting a multi-cloud strategy. This approach distributes workloads across several providers to prevent any one failure from crippling operations.

It can be further strengthened by “edge computing,” which decentralises processing and storage, moving them closer to users through smaller, local servers. By combining edge computing with a multi-cloud setup, businesses can enhance resilience, improve speed, and maintain greater control over data.

As the old saying goes, don’t put all your eggs in one basket — especially when the basket belongs to someone else.

The Conversation