The robot vacuum did not start. Doorbell cameras have stopped monitoring package thieves, but some of their deliveries have been canceled anyway. The Netflix and Disney movies were interrupted, and the Associated Press had problems publishing news.
Amazon’s cloud computing network went down on Tuesday, causing significant disruption to the services of various U.S. companies over the course of several hours, questioning Internet vulnerabilities and a focus on a small number of companies. it was done.
How did it happen?
Amazon hasn’t said exactly what went wrong yet. The company restricted communications on Tuesday, simplified the technical description of the Amazon Web Services dashboard, and through spokesperson Richard Rocha admitted that the outage affected Amazon’s own warehousing and delivery operations. In a short statement, the company “is working to resolve the issue quickly. As much as possible.” We didn’t immediately answer further questions on Wednesday.
The Amazon Web Services incident primarily affected the eastern United States, but nevertheless affected everything from airline bookings and car retailers to payment apps and video streaming services to Amazon’s own large-scale e-commerce. I did.
What is AWS?
Amazon Web Services is a cloud service operation that stores customer data, performs online activities and more, and is Amazon’s huge profit center. According to research firm Gartner, the company occupies about 40% of the $ 64 billion global cloud infrastructure market, with a larger share than its closest rivals Microsoft, Alibaba and Google combined.
Previously, it was run by Amazon CEO Andy Jassy, who took over the founder Jeff Bezos in July.
Are there too many eggs in one basket?
Some cybersecurity experts have been warning for years about the potentially ugly consequences of allowing a handful of large tech companies to dominate major Internet operations.
“The recent AWS outage is a classic example of the dangers of centralized network infrastructure,” said Sean O’Brien, a visiting lecturer at Yale Law School’s cybersecurity. “Most people who browse the internet or use apps don’t know that, but Amazon is built into most apps and websites they use every day.” O’Brien said early Internet peer-to-peer. He said it is important to build a new network model that resembles the roots of. Massive outages have already knocked out vast areas of the world offline, as happened in the Facebook incident in October.
Even with the current model, enterprises have several options for splitting services between different cloud providers, but they can be complicated. Or, at the very least, you can allow the service to be moved to another region operated by the same provider. The outage on Tuesday primarily affected Amazon’s “USEast 1” region.
“That is, if we had a critical system that was only available in that area, we had a problem,” said Servaas Verbiest, lead cloud evangelist at SunGard Availability Services. “If you’re very embraced by the AWS ecosystem and stick to using only its services and features, you need to make sure you balance your workload across regions.”
Has this ever happened before?
Yes. The last major outage for AWS was November 2020. There have been many other destructive and long-term Internet outages involving other providers. In June, behind-the-scenes content distributor Fastly failed, temporarily shutting down dozens of major internet sites, including CNN, the New York Times, and the UK Government’s home page. Another month of the month affected provider Akamai during peak Asian business hours in June.
With the October outage, Facebook (now known as Meta Platforms) has taken down Instagram and WhatsApp in addition to its nominal platform, a “wrong configuration change” of an hour-long global outage. Blame.
What about the government?
It was unclear how or if Tuesday’s outage affected the government, but many governments also depend on Amazon and its rivals.
Among the most influential organizations rethinking the approach of relying on a single cloud provider was the Pentagon. The Pentagon canceled its pending cloud computing contract with Microsoft in July. Instead, we will pursue transactions with both Microsoft and Amazon, and perhaps with other cloud service providers such as Google, Oracle, and IBM.
Earlier this year, the National Security Agency awarded Amazon a potential estimated $ 10 billion contract to be the only manager of NSA’s own migration to cloud computing. This agreement is known by the agency codename “Wild and Stormy”. The October General Accountability Office upheld Microsoft’s bid protest and determined that certain parts of the NSA’s decision were “irrational,” but the full decision has been categorized.
Explainer: What is the cause of Amazon’s outage? Do you have more? Chicago News
Source link Explainer: What is the cause of Amazon’s outage? Do you have more? Chicago News