Language
Country/Area
No result found
The secret of serverless computing: Why are servers still working silently in the background?
With the development of cloud computing, serverless computing has attracted the attention of more and more enterprises. However, behind this concept lies the secret that many servers are still quietly helping. According to the definition of ISO/IEC 22123-2, serverless computing allows customers to use cloud services without configuring, deploying or managing hardware or software resources, and only needs to provide application code or data.
The term serverless is not entirely accurate because cloud service providers still need servers to execute the developer's code.
The existence of servers changes the definition of serverless computing. Ben Kehoe proposed that serverless computing should be viewed more as a spectrum rather than a rigid definition. This means that enterprises should focus on leveraging the benefits of serverless solutions rather than being restricted to a narrow definition of a specific technology. Although serverless computing simplifies the process of code deployment, according to Gregor Hohpe, it does not completely eliminate complexity, but moves it from the operations team to the development team.
Serverless code can be used with traditional styles such as microservices or monolithic architecture, or even without relying on any configured server at all. This should not be confused with certain computing or network models that do not require an actual server to run, such as peer-to-peer (P2P) models. According to Yan Cui, serverless computing should be adopted when it can help deliver customer value faster.
Serverless runtime
The computing runtime provided by serverless vendors can execute application logic but is not responsible for data storage. Common runtime models include Function as a Service (FaaS) and Container as a Service (CaaS). Common supported languages include Java, Python and PHP.
Commercial Products
The history of serverless computing can be traced back to Zimki, launched in 2006, but the platform was not successful. Subsequently, Google launched Google App Engine in 2008, which was the first module to provide billing applications.
AWS Lambda was launched by Amazon in 2014, further popularizing the abstract model of serverless computing.
Serverless databases such as Amazon Aurora and Azure Data Lake have also successfully extended this computing model to relational database management systems (RDBMS), eliminating the need for configuration and expansion of physical or virtual database hardware. This shift allows enterprises to focus more on their business logic rather than infrastructure.
Advantages of serverless computing
Serverless computing offers multiple advantages, especially in terms of cost-effectiveness and flexibility. This computing model avoids the high expense of leasing or handling fixed servers and automatically expands as demand changes. Development teams are able to run code quickly without digging into infrastructure, allowing them to focus more on business logic and delivering customer value.
Challenges of serverless computing
However, serverless computing is not without its challenges. Response delays, resource constraints, difficulties in monitoring and debugging, and security concerns are all factors that enterprises need to carefully consider when adopting serverless solutions.
Some brute force attack or defective code could cause users to experience unexpected payment fluctuations, which is the risk of "bill shock".
In addition, relying on serverless services provided by third parties can also lead to vendor lock-in issues, which makes enterprises face difficulties when migrating workloads.
Best Practices
Adopting DevSecOps practices can help enterprises use and protect serverless technologies efficiently, further blurring the boundaries between development and security, and making security an important component of multi-party cooperation.
Although this emerging computing model can technically solve many problems, in actual business applications, companies still need to carefully consider it and not ignore potential risks because of convenience. In the face of such a rapidly developing technology field, are enterprises ready to meet the challenges and opportunities brought by serverless computing?
