Serverless
architecture or function as a service, FaaS, is a design pattern where applications
or programs are hosted in cloud service. Scaling, capacity planning and server maintenance
are hidden from the developers. Applications are typically divided into small,
short lived functions where a function is responsible for a single task. This
makes it easy to scale different parts of the applications if necessary.
Serverless Integration
Serverless approach does almost always require event-based architecture. Functions (jobs) are designed to handle a single task and then deployed in a cloud. The functions are triggered by some external event, like http-request, file creation or by timer. All this fits nicely into how integrations typically are built up.
However when running serverless you don’t have access to you on-premises resources which heavily limits the use of this approach in integration jobs. But if we could access some local resources like a filesystem or a database, running serverless could be an option at least for jobs that has a bit longer running time. Moving data between on-premises resources with a serverless-integration-job would certainly add execution time in form of data transfer and that’s idle computing power in the cloud that you need to pay for. Therefor jobs that process the data or do some type of calculations are preferable thus the impact of the data transfer will be smaller.
How can Netlang help you with serverless integration?
- Install our on-premises Gateway and securely open up your local resources to the cloud. Define which resources that should be available in your Netlang Cloud and enjoy cloud benefits while your data is stored on-premises.
- Run your integration jobs serverless in the cloud and use your local resources. All jobs are location independent and may be run on-premises or in the cloud.
- Define your execution environments and let Netlang determines where your integration jobs should be run. Netlang analyses your code and use Machine learning to determine where your jobs should be run, based on what you prioritize, speed, cost or …
- Open your local data resources to your partners. In Netlang you can collaborate with your business partners and share on-premises resources.
