Why Crypto’s Infrastructure Still Lags Behind Its Vision

Frequent outages and the growing pressure placed on major cloud providers have created a clear opportunity for companies developing decentralized networks.

Advocates of this distributed model argue that dividing workloads across many smaller nodes lowers the risks associated with centralized control. They believe this structure could be especially useful in fields where computing power is heavy and interruptions are unacceptable—think AI, gaming, and financial systems.

“As decentralized infrastructure continues to reach, and eventually surpass, the capabilities of traditional cloud services, dependence on single providers will naturally fade,” said Carlos Lei, CEO and co-founder of Uplink, a DePIN-powered connectivity marketplace, in a statement to Cointelegraph.

In modern computing, decentralized infrastructure typically points to blockchain technology, which is built to spread trust, prevent bottlenecks, and eliminate single points of failure by dispersing data storage and verification.

Yet, despite these goals, most of the systems that give users access to blockchain networks still operate on centralized cloud services—revealing a gap between crypto’s ideals and the realities of its current infrastructure.

Reliance on the Big Three—and the Risks of Centralized Cloud Power

As of 2024, Amazon Web Services (AWS), Microsoft Azure, and Google Cloud—collectively known as the “Big Three”—controlled roughly 68% of global cloud infrastructure revenue, according to Synergy Research Group. Their dominance has made them the default backbone for countless enterprise platforms and blockchain-related services.

However, this level of concentration comes with a downside: a single outage, sudden pricing shift, or regional capacity issue at any one of these giants can disrupt entire industries.

“When AWS went down in October, platforms like Snapchat, Roblox, Fortnite, and even Kindle were completely offline,” said Nökkvi. Dan Ellidason, CEO of the tech infrastructure firm Gaimin, in an interview with Cointelegraph. “Coinbase—a financial service—was hit hard as well.”

These major cloud providers have long offered startups substantial support by lowering initial infrastructure costs. But stepping away from centralized cloud environments means reclaiming the responsibility for configuration, security, and upkeep—tasks the Big Three have handled for years—and absorbing the higher expenses that come with it.

Even so, a number of companies are already experimenting with alternative setups through gradual, modular transitions. Data and metadata are increasingly being stored on solutions like Filecoin and Arweave, while decentralized compute and GPU power are emerging through networks such as Akash and Render.

Centralized Clouds Are Quietly Weakening Blockchain’s Decentralization Goals

Blockchains are built to distribute trust, reduce fragility, and avoid single points of failure—yet much of the infrastructure supporting them still sits on centralized cloud platforms. In many major proof-of-stake networks, a significant portion of validators rely on commercial cloud providers instead of independent or self-hosted hardware.

Research presented at the Usenix Security Symposium in August highlighted this imbalance, identifying Amazon Web Services as the largest hosting provider among the Ethereum validators the researchers were able to map—roughly 20% of the total. Likewise, a 2023 report from Messari showed that most validators and staking services operate on centralized servers.

“Many operators pick platforms like AWS or Google Cloud because they offer consistent performance, straightforward configuration, and dependable uptime,” Lei explained.

“These services have effectively become the foundation for much of Web3 because they remove the operational pain of scaling and managing infrastructure.”

Users rarely notice this heavy dependence—until something breaks. During the AWS outage on Oct. 20, Coinbase customers struggled to log in, execute trades, or withdraw assets, while Robinhood traders faced delays and API failures.

Hybrid Cloud Models Emerge as Blockchain Expands

The increasing pressure on centralized cloud providers is driving a new generation of infrastructure projects to experiment with distributed alternatives. These emerging networks tap into consumer hardware, regional data centers, and unused computing resources to build more resilient systems.

Gaimin is one example, pooling GPU power from gaming computers and reinforcing it with capacity from smaller regional data centers. According to Ellidason, this design creates a geographically diverse network that avoids having any single region become a critical point of failure.

Uplink follows a similar decentralized logic, but with connectivity instead of computing. Its marketplace allows individuals and local operators to sell surplus bandwidth to applications requiring broader coverage, helping reduce dependency on large telecom providers.

Many researchers and node operators believe the future won’t be fully decentralized but hybrid—blending hyperscalers, edge networks, and bare-metal machines, with automated failover when one region experiences disruption.

As demand for compute continues to exceed the capacity of major cloud platforms, the shift isn’t about abandoning AWS or other big providers. Instead, it’s about enhancing resilience and opening opportunities for regional infrastructure players to cover gaps that hyperscale data centers can’t always address.