FaaS, the next step towards a serverless cloud
FaaS, Functions-as-a-Service , could become the highest level of abstraction for cloud services. The major players in the cloud and cloud services are seeking to evolve their technologies to ever higher levels of abstraction. It all started with the sharing and reselling of virtualized infrastructure, Infrastructure-as-a-Service ( IaaS ).
Then progress in sophistication with analytics, machine learning, network services, or development and application backends resulted in PaaS ( Platform-as-a-Service ). That’s good, publishers today unveil platform strategies.
However, there is a higher echelon in the logical abstraction ladder of the cloud-based support infrastructure, which giants like AWS or Microsoft call Serverless Computing, or FaaS ( Functions-as-a-Service ). .
The naming is clumsy, code services like AWS Lambda, Azure Logic Apps, or Google Cloud Functions always run on servers and physical architectures, usually in Docker-type containers. However, for the user, serverless means that it does not need to worry about the details of deployment, server monitoring, or resource sizing.
FaaS is therefore the furthest approach from the implementation details, turned up for valuable services, which however remains subject to a broad interpretation. According to Microsoft, this is a concept identified by several important attributes:
- Complete abstraction of the user’s servers
- Event-driven services that are dynamically and instantly scalable;
- A pricing model for micro-billing services based on the consumption of resources and executions, not on the size of the instances.
The advantage is obvious for developers, which removes the need to worry about the infrastructure and increases the time to focus on the business logic of an application. With the reduction of DevOps processes and shorter cycles. However, Faas is not suitable for all issues, and some modules need to integrate with other programming models that require persistent PaaS and IaaS.
- Above, an example of FaaS via Logic Apps from Microsoft, an Azure version of BizTalk Server that provides the orchestration of the serverless workload between data sources and other cloud computing applications and services.
Even if they are only in the early stages of analysts ‘or marketers’ expectations, enterprise integration services are not new. On the other hand, they integrate concepts like sophisticated data analysis, machine learning, and cognitive services for things like predictive statistical modeling, image, or speech recognition by making them accessible to non-developers.
Organizations that learn to properly integrate functions into a cloud architecture with no application server (FaaS), will find a powerful microservice module development tool that offers significant benefits, to complement, not replace, application containers and more advanced cloud platforms for data and artificial intelligence services.