In the past three weeks, two major power outages in Amazon's cloud computing services have caused widespread outages in other online services. Last month, a problem with Comcast, one of the largest Internet service providers in the United States, caused widespread outages. In June of this year, when cloud computing service provider fast was dealing with the issue of "service configuration", websites across the globe were temporarily offline.
The most recent outage occurred on Wednesday when users of DoorDash, Hulu, and other sites complained that they could not connect. These problems have been traced back to Amazon Web Services (AWS), the most widely used cloud service company. The company reported that outages in two of its 26 geographic regions affected services across the United States.
The most recent outage occurred on Wednesday when users of DoorDash, Hulu, and other sites complained that they could not connect. These problems have been traced back to Amazon Web Services (AWS), the most widely used cloud service company. The company reported that outages in two of its 26 geographic regions affected services across the United States.
Over the past decades, the speed and reliability of the Internet have continued to increase, strengthening people’s illusions that everyday consumers can rely on online services without fail. These events helped break this illusion.
Robert Blumofe, executive vice president and chief technology officer of Akamai Technologies, said that in the past, an online video meant watching "5 minutes of low-resolution video." Akamai sells security services and "edge computing" functions, a distributed technology that does not rely heavily on centralized data centers.
Blumofe said: "Now, people are looking forward to watching the entire movie in high resolution." "This is a recent prejudice. We remember the situation then and now, but not the past."
In other words, some Americans who enjoy reliable Internet access may be a bit spoiled.
Experts in computer science and security say that these interferences do not make people question the basic design of the Internet. One of the founding ideals of the Internet is that even if one component fails, the distributed system can continue to operate to a large extent. But they said that the root of these problems lies in the unbalanced development of the Internet because some data centers are more important than others; the cloud business operated by Amazon, Google, and Microsoft has concentrated more power. Enterprise customers of cloud services are not always willing to pay extra for backup systems and employees.
Sean O'Brien, a cyber security lecturer at Yale Law School, said the outages have raised questions about the wisdom of relying so much on big data centers.
He said: "The'cloud' has never been sustainable. It is just a euphemism for a centralized network resource controlled by a centralized entity." He added that alternative technologies such as peer-to-peer technology and edge computing may be favored. After the outage last week, he wrote that the large cloud provider is simply a "feudal" system.
Cloud service providers make money by selling server space to other companies with flexible terms and expertise, reducing the need for companies to manage their servers. They rarely fail, but when they fail, they get attention. The AWS downtime in November 2020 affected customers such as Apple.
Vahid Behzadan, assistant professor of computer science at the University of New Haven, said: "There are many points of failure, and their unavailability or poor operation will affect the entire global Internet experience."
Some of these mistakes-such as the AWS "us-east-1" area-have become notorious among tech workers who share their experience on industry message boards.
Bezadan said: "We have been interrupted many times in a short period of time. This is a fact worthy of vigilance." He pointed out that US companies have made a big bet on the assumption of cloud service elasticity.
He also said that if outages become more common or publicly visible, corporate customers may spend more money on backup systems to ensure that they are resilient in the event of failures—for example, they have contracts with Google and Amazon. According to the National Broadcasting Corporation Financial Channel (CNBC), the industry debate about whether to go "cloudy" is now rekindled, and companies in various industries are investing more in edge computing tools.
"The Internet will not die in the short term. But anyway, as long as it can't kill the Internet, it will become stronger."
Robert Blumofe, executive vice president and chief technology officer of Akamai Technologies, said that in the past, an online video meant watching "5 minutes of low-resolution video." Akamai sells security services and "edge computing" functions, a distributed technology that does not rely heavily on centralized data centers.
Blumofe said: "Now, people are looking forward to watching the entire movie in high resolution." "This is a recent prejudice. We remember the situation then and now, but not the past."
In other words, some Americans who enjoy reliable Internet access may be a bit spoiled.
Experts in computer science and security say that these interferences do not make people question the basic design of the Internet. One of the founding ideals of the Internet is that even if one component fails, the distributed system can continue to operate to a large extent. But they said that the root of these problems lies in the unbalanced development of the Internet because some data centers are more important than others; the cloud business operated by Amazon, Google, and Microsoft has concentrated more power. Enterprise customers of cloud services are not always willing to pay extra for backup systems and employees.
Sean O'Brien, a cyber security lecturer at Yale Law School, said the outages have raised questions about the wisdom of relying so much on big data centers.
He said: "The'cloud' has never been sustainable. It is just a euphemism for a centralized network resource controlled by a centralized entity." He added that alternative technologies such as peer-to-peer technology and edge computing may be favored. After the outage last week, he wrote that the large cloud provider is simply a "feudal" system.
Cloud service providers make money by selling server space to other companies with flexible terms and expertise, reducing the need for companies to manage their servers. They rarely fail, but when they fail, they get attention. The AWS downtime in November 2020 affected customers such as Apple.
Vahid Behzadan, assistant professor of computer science at the University of New Haven, said: "There are many points of failure, and their unavailability or poor operation will affect the entire global Internet experience."
Some of these mistakes-such as the AWS "us-east-1" area-have become notorious among tech workers who share their experience on industry message boards.
Bezadan said: "We have been interrupted many times in a short period of time. This is a fact worthy of vigilance." He pointed out that US companies have made a big bet on the assumption of cloud service elasticity.
He also said that if outages become more common or publicly visible, corporate customers may spend more money on backup systems to ensure that they are resilient in the event of failures—for example, they have contracts with Google and Amazon. According to the National Broadcasting Corporation Financial Channel (CNBC), the industry debate about whether to go "cloudy" is now rekindled, and companies in various industries are investing more in edge computing tools.
"The Internet will not die in the short term. But anyway, as long as it can't kill the Internet, it will become stronger."
