Tencent Cloud Serverless Cloud Function (SCF) is a serverless execution environment that enables you to build and run applications without having to purchase and manage servers. It is an ideal computing platform for use cases such as real-time file processing and data processing. Simply code in a supported language and set the execution conditions, and your code can be run on the Tencent Cloud infrastructure elastically and securely.
Evolution of Computing Resources
With the development of cloud services and the high abstraction of computing resources, Tencent Cloud provides a wide variety of computing resource options from physical servers to cloud-based functions at different abstraction levels.
Cloud Bare Metal (CBM): scaling is at the physical machine level. You enjoy the physical computing resources exclusively, which provides the best security. Cloud Virtual Machine (CVM): it virtualizes hardware devices, and scaling is at the virtual machine level. You share the physical machine resources with other tenants, and you can configure CVM metrics on your own, which is simpler than deployment and iteration. Tencent Kubernetes Engine (TKE): it virtualizes operating systems, and scaling is at the service level. The test and production environments are exactly the same, making testing and deployment very convenient. Serverless Cloud Function (SCF): it virtualizes runtime environments, and scaling is at the function level. This is the smallest unit of existing computing resources, which features full automation, one-click deployment, and high scalability, and is an ideal choice for lightweight service deployment. Serverless
Serverless does not mean that there is no server. It is just that you do not need to care about the underlying resources, log in to a server, or optimize the server. You only need to care about the core code snippets while skipping the complex and cumbersome basic work. These code snippets are completely triggered by events or requests, and the platform automatically adjusts service resources in parallel based on the requests. The serverless architecture has near-infinite scalability, and no resources will be executed when idle. The code is executed in a stateless manner, which easily enables fast iterations and rapid deployment.
SCF Overview
SCF is a serverless execution environment provided by Tencent Cloud. All you need to do is write simple and single-purpose functions and associate them with events generated by your Tencent Cloud infrastructure and services.
When using SCF, all you need to do is write the code in a programming language (Python, Node.js, PHP, Go, Java, and Custom Runtime) supported by the platform. The underlying computing resources and tasks are fully managed by Tencent Cloud, including maintenance of server CPUs, memories, networks, and other configurations/resources, code deployment, elastic scaling, load balancing, security upgrade, and resource execution monitoring. However, this means that you cannot log in to or manage servers or customize the system or environment.
SCF is automatically deployed in multiple AZs in the same region with extremely high fault tolerance achieved. When a function is executed, scaling will be made based on the request load with no manual configuration or intervention required, helping meet the needs for service availability and stability in different scenarios. From several requests per day to thousands of requests per second, SCF can automatically scale at the underlying layer. You only need to pay for running functions, and if no function is running, no fees will be incurred.
You can customize the timing of running a function, such as when a file is uploaded to a COS bucket, when a file is deleted, when a message in CKafka is used, or when an application is called through an SDK. You can also configure the function to run regularly. Therefore, SCF can be used as a data processing trigger for the COS service to easily implement IFTTT logic. Scheduled automated tasks can also be flexibly built to free you from manual operations and easily construct an elastic and controllable software architecture.
SCF Features
Serverless helps you get rid of the tedious development and configuration work, so that you can only care about the writing of business code logic, without any infrastructure construction, management, and OPS overheads. This service model lowers the threshold of R&D and improves the efficiency of business construction; therefore, it has gained the recognition of high numbers of enterprises and developers.
Various development tools and languages supported for smooth development
The Tencent Cloud Serverless team works in many ways to provide convenient tools or capabilities that can meet the needs in a wide variety of development scenarios; for example:
With the aid of visual operations based on the VS Code plugin and IDE, online/offline management of functions and code writing and debugging can be performed at one single place. The VS Code IDE and plugin also enables local management, development, and debugging as well as online release of functions.
Web IDE is supported; therefore, you can develop and debug projects in real time in the console, which delivers the same experience as local development and debugging and makes it easier for you to view and adjust code. SCF supports Python, Node.js, Go, PHP, Java, and Custom Runtime, so you can select a custom runtime environment as needed. Multiple deployment methods for various environments
SCF supports deployment through the console, command line, SDK/API, web IDE, and image.
Diversified triggers to support more business scenarios
Triggering methods of SCF include API, SDK, and events in other Tencent Cloud services such as COS and API Gateway, which enrich the use cases.
Automated and flexible execution for better invocations
SCF can automatically scale according to the call volume, which is imperceptible to users, perfectly fits the invocation curve, and saves resources and costs to the greatest extent.
Pay-as-You-Go billing accurate to the millisecond level
SCF supports billing actually used resources at a time granularity of 1 ms, which can significantly reduce your costs compared to the time granularity of 100 ms.