亞馬遜雲端運算崩潰：當網路暫停時

Amazon Web Services, known as AWS, experienced a major outage on Monday that disrupted many popular websites and online services. AWS is a cloud computing platform owned by Amazon, and it helps run websites and apps for companies all over the world. When AWS had technical issues, sites like Snapchat, Reddit, and Lloyds Bank were affected and could not function properly. Users trying to access these platforms saw error messages, and some services were unavailable for several hours. Amazon issued an apology to all customers impacted by the outage, acknowledging how important their services are to businesses and individuals. They said they understand the critical role AWS plays in supporting applications and end users, and they are committed to improving their systems to prevent similar problems in the future.

The outage was traced back to AWS's largest group of data centers in North Virginia, United States. These data centers are responsible for connecting websites to the internet by matching website names with the IP addresses that computers use. The problem started when errors in Amazon's internal systems prevented websites from being linked to their correct IP addresses. This was caused by a technical issue in the automated processes that manage the Domain Name System, or DNS, which helps computers understand website URLs. According to Amazon, a dormant bug was triggered by a delay in one of these processes, causing the systems to fall out of sync and stop working properly.

While some platforms like Roblox and Fortnite were restored within a few hours, others experienced longer periods of downtime. Lloyds Bank customers, for example, faced issues until the afternoon, and services like Venmo and Reddit were also affected. The outage even impacted smart beds made by Eight Sleep, which rely on internet connections to adjust temperature and elevation. Experts pointed out that the incident highlights how much the tech industry depends on AWS and similar cloud providers. Dr Junade Ali, a software engineer, explained that faulty automation was at the core of the problem, and he suggested that companies should use multiple cloud providers to avoid being taken offline by a single failure. Amazon said they will do everything possible to learn from the event and improve the reliability of their services.