The AWS outage is proof that a decentralized web isn’t just desirable — it’s essential | Opinion

Blockchain developers have been banging on for years about the need to decentralize the web with the goal of mainstreaming more robust, shock-resistant systems. Decentralized infrastructure, they insist, offers greater redundancy by eliminating reliance on the monopolistic providers that currently form the backbone of the internet.

Compelling as this argument is, it struggles to gain much attention outside of blockchain circles so long as the web is working as it should. But the world received a wake-up call on October 20 about the dangers of centralized systems with the AWS global outage that impacted core services for hundreds of millions of web users, from consumers to businesses. While the costs of the incident are still being counted, they’re projected to run into hundreds of billions of dollars.

If the notion of the decentralized web once attracted the interest of global enterprises, now it has their attention. It is an idea whose time has come. But what might such a model resemble in practice, and could an internet built on decentralized infra provide fault-tolerance while still meeting the needs of data-hungry digital companies? Let’s examine the evidence.

The case for decentralizing the web

The current web infrastructure depends heavily on a handful of providers like AWS, which control vast portions of cloud services, meaning that when vulnerabilities occur, the effects are cataclysmic. The AWS outage of October 20 disrupted services for millions, including major apps and businesses, revealing single points of failure.

The problem stemmed from a specific AWS misconfiguration — DNS issues in the company’s US-EAST-1 region, to be precise — but its effects were felt globally due to Amazon’s dominance of the cloud market. It’s understandable why enterprises should rely on AWS, given that most of the time its service is both reliable and scalable, but these very attributes are responsible for making its cloud service a liability — an accident waiting to happen.

With AWS holding over 30% of the global cloud market, this concentration amplifies risks during failures. This isn’t about what happens during 99.9% of the time when the cloud is working as expected — it’s about what happens the other 0.1% of the time, when edge cases are elevated into systemic shocks.

But it didn’t have to be this way. On an internet whose infrastructure is blockchain-based, such incidents would be virtually impossible because there is no single point of failure. Indeed, it’s hard to envisage a scenario in which blockchain architecture, in which data is stored across numerous nodes rather than centralized servers, could succumb to the same fate as beset Amazon.

Peer-to-peer networks are expressly designed to ensure that no single party can unilaterally wield control. This is about more than merely preventing centralization, however: it ensures fault-tolerance. If a single node is knocked offline — or even if an array of nodes somehow succumbs — the network will keep ticking over. No panic. No downtime.

The web, distributed

It’s easy to paint broad brush strokes outlining the benefits of a decentralized web, but what might this look like in practice? In real terms, it would entail placing the primary resources that power the web on blockchain rails, including decentralized storage and compute. Whether it’s running AI workloads using decentralized physical infrastructure networks to supply GPU clusters or P2P protocols that provision terabytes of storage, web3 is more than capable of delivering.

We often envision these core blockchain components in the context of censorship resistance, and while this framing holds true, it equally describes shock-resistance: if one component fails, the network remains intact and globally accessible.

In reality, it’s unrealistic to expect enterprises to switch to decentralized infra overnight just because AWS had a bad day. But it is realistic to project that forward-thinking businesses will look to test the waters and see what distributed web services have to offer, slowly at first and then with haste once they discover that decentralized infra is up to the task.

Initially, such enterprises are likely to adopt a hybrid approach, combining centralized cloud services with blockchain equivalents for backend infrastructure, giving them a failsafe that’s always available and ever dependable. But if web3 can prove its ability to meet global demands, uptake will accelerate.

Maximum uptime, minimum downsides

The decentralized web has escaped the lab of web3 engineers, crafting big brain whitepapers and theorizing nodes that can solve the Byzantine Generals Problem. It’s now in production and seeing real-world adoption — adoption that’s poised to heighten in the wake of this week’s seismic events.

Greater resiliency is the primary selling point, given the ability for nodes to reroute traffic automatically, circumventing Amazon’s nine-hour outage that brought the internet to its knees. But there are other benefits of transitioning to decentralized infra, including the cost savings it can engender.

Eliminating intermediary fees and expensive cloud dependencies in favor of P2P alternatives can reduce spending by up to 90%. And then there are the security benefits that decentralized infra brings to the table. Through distributing data by splitting it into pieces and storing it across thousands of nodes, hackers have no central repository to target.

The tech is ready

In the past, the concept of a decentralized web was a vision more than a working reality since the tech would have struggled to scale as smoothly as centralized systems had enterprises suddenly switched in their droves.

Not any more. The latest generation of web3 infra is faster, cheaper, and more reliable than anything that’s gone before and is capable of giving the leading cloud providers a run for their money. Crypto networks like Bitcoin (BTC) have operated with 100% uptime for over a decade, while DePIN has shown that enterprises can have their needs met by decentralized networks that connect them with the resources they need to thrive.

If Bitcoin is the money of last resort, blockchain is the infrastructure of last resort. Not because that’s the only occasion when it should be utilized, but because such occasions are precisely when it shines. You don’t need a decentralized web until you do.

As the old political maxim goes, “Never let a crisis go to waste.” Make no mistake, the AWS outage was a crisis that was waiting to happen. On October 20, it detonated with far-reaching consequences, the aftershocks of which are still being felt. On this occasion, it was “just” AWS that was toppled. Imagine what would happen if the gremlin in the works also took out Azure and Google Cloud? The internet as we know it would grind to a halt.

The world has woken up to the dangers of relying on centralized infrastructure. The onus is now on the blockchain industry to capitalize on this crisis, not to score points, but to demonstrate that there’s a better way to build the web – one that will keep core services online all the time.