Continuing from the last decade, we have observed a major innovation in the enterprise IT space. The primary focus of these innovation has been on providing business agility, improving latency and drive cost effectiveness. If we talk about server computing, the units of scaling are becoming smaller, for example, from virtual machines to containers to serverless.
In the traditional environment, companies were required to pay a fixed and a recurring amount for using virtual machines to run their websites irrespective of whether their cloud servers are being used or not. With serverless architecture, you only need to pay only for the service usage without any upfront charges i.e. pay-per-execution. In and all, serverless computing is an evolution of cloud computing service models, from IaaS (infrastructure-as-a-service) to PaaS (platform-as-a-service) to FaaS (function-as-a-service).
Leading cloud providers like Amazon, Microsoft, Google & IBM have all launched serverless services in the last 5 years. Amazon’s service is called AWS Lambda while Microsoft and Google are called Azure Functions and Google Functions respectively.
Current status of serverless landscape
Since the beginning of the year 2018, four major reports have made it evident that serverless computing is growing at an incredible pace which makes it one of the hottest cloud-related technology. This is mainly because the leading providers are enhancing their serverless offerings with add-on features. Also, the community and serverless use cases are growing which is accelerating the adoption of these technologies.
To present you the gist of the reports, let’s discuss some of the key findings from all these reports:
- With the growth rate of 75%, serverless computing is the number one cloud services as reported by the RightScale 2018 State of the Cloud report.
- Currents 2018 from DigitalOcean reported that the most beneficial aspect of serverless computing is the reduction in costs related to the server provisioning. While the biggest obstacle in the adoption of serverless technology is the difficulty in monitoring and debugging.
- Cloud Foundry 2018 reported that 46% of surveyed companies are currently using serverless computing.
Changing the technology stack
The onset of serverless technology is also redefining the ways in which we have been developing our software architecture. To be specific, it is actually changing the traditional technology stack of the network, compute and database we’ve been using so far. The three key core technology components of serverless computing stack include the following:
- Network = API Gateway
In serverless architecture, the API Gateway acts as a communication layer between the frontend and the serverless functions (FaaS) layer. It maps REST API endpoints with the corresponding functions that run the business logic. With servers out of question, there is no requirement for deploying and managing the load balancers also in this model.
- Compute = Function-as-a-Service
This is a layer which executes a particular business logic with the help of small code blocks. Though FaaS requires the level of abstraction in terms of executing the business logic provided by cloud providers like AWS, Azure & GCP.
- Database = Backend-as-a-Service
This is essentially a cloud-based distributed NoSQL database which essentially removes database administration overheads. In the current status, it is widely used for authentication services.
The serverless landscape is constantly changing with the new updates and releases of new products. The new areas of services are being tapped sooner. However, the winners will be defined by how seamlessly they can incorporate these changes in their traditional solutions. In this matter, the current reports depict the bright future. Though whether it becomes a reality or not, only time will tell.