Barely a week after Amazon Web Services (AWS) suffered a major outage affecting customers around the world, on Wednesday, Oct. 29, Microsoft Azure experienced a similar cloud failure.
The recent outages have reignited calls for blockchain-based alternatives, which advocates argue are more resilient than centralized cloud services.
Starting at around 15:30 UTC on Wednesday, reports of Azure outages on DownDetector spiked, remaining elevated for around two hours. Other services were also affected, including Microsoft 365 and Xbox Live.
In a statement on Azure’s live status page, Microsoft said the issue was related to its application delivery network, Azure Front Door. Blaming the incident on an “inadvertent configuration change,” the firm said it managed to restore services by rebooting the system to its last-known working state.
Despite Microsoft’s actions to mitigate the problem, some users reported disruption lasting 6–8 hours, a similar timeframe to the previous AWS failure.
After the AWS outage caused service disruption for Coinbase, Snapchat, Fortnite, and others, the Azure blackout wreaked even more havoc. Around the world, Starbucks, Vodafone, Heathrow Airport, and Natwest were among those affected.
According to the latest research from Synergy, the Big Three hyperscalars—AWS, Microsoft, and Google—account for nearly two-thirds of the global cloud market. In Q3 2025, AWS alone had a 31% market share.
Over the years, incidents of downtime have done little to slow cloud adoption or the growth of Big Tech services. Nonetheless, recent outages prompted some companies to reevaluate how much they rely on them.
For instance, the smart bed manufacturer Eight Sleep shipped a new “outage mode” after the temperature control feature on its mattresses started behaving erratically while AWS was down.
Despite criticism of the hyperscalar cloud model, or at least, a recognition of its vulnerabilities, business leaders don’t all agree on the best remedy.
In the wake of recent outages, systems engineers around the world have been debating the merits of multi-cloud versus hybrid or on-premise solutions. Meanwhile, for proponents of decentralization, blockchains offer a compelling solution to many cloud challenges.
Although it’s hard to capture the full range of services under a single, overarching definition, a few key themes characterize the decentralized cloud movement.
Storage is arguably the most established layer of the decentralized cloud stack. Platforms like IPFS, Filecoin, and Arweave connect hardware nodes in a distributed network, using blockchain ledgers to record where data is stored and orchestrate payments.
In recent years, GPU networks like Render and Aethir have applied the same logic to compute. As with decentralized storage solutions, these platforms use blockchains to connect buyers and sellers, each with their own tokenomic model.
Finally, several projects are developing full-stack decentralized clouds that combine storage and compute. For instance, Internet Computer (ICP) “canisters” function as smart contracts with memory and compute capacity, enabling applications that run fully on-chain.
While blockchain networks can be more resilient than centralized servers, decentralization isn’t a silver bullet. It also comes at the cost of increased latency, which slows delivery and drives up costs.
Axelar co-founder Sergey Gorbunov has argued that the best way to apply blockchains to the cloud market is as a neutral orchestration layer. In this model, cloud providers of various types compete in an open market. If one goes down, the blockchain automatically routes users to a live alternative.
Gorbunov’s vision promotes fairer and more transparent pricing. This stands in contrast with the current cloud market, in which hyperscalers lock customers into fixed contracts that make switching providers difficult.