Which Applications Are Best Suited for Serverless Architecture?
Serverless can be a cost-effective computing environment if the service is rarely used. That's because agencies only pay for resources they use. If the service is constantly spun up, the costs can add up quickly.
There are also limits to how far a serverless architecture can go in terms of elasticity. If an agency has a larger application that exceeds the resource limits set by the cloud provider, serverless may not be a good fit.
Because of this, serverless architectures can serve an agency well in cases where an application is designed to run code and perform trigger-based tasks. While waiting for a trigger event, the application remains passive on the cloud infrastructure and doesn’t use up more resources, keeping costs low.
Continuous integration/continuous delivery (CI/CD) is another use case. In a serverless environment, developers can constantly update production code without updating the server. Serverless literally abstracts those developed processes away. This real-time ability to update code can significantly speed service delivery.
EXPLORE: How edge computing can deliver better application performance.
Serverless Architectures Can Bring Complexity
Reducing the cost of maintaining an IT architecture while becoming more agile is an attractive proposition for any agency, but serverless also introduces complexity.
First, since the cloud provider owns and oversees the backend, it is almost impossible for an agency to monitor the applications running behind the scenes. Due to this limited visibility, developers may have to guess how their application is performing.
Using a variety of monitoring tools is an option, but this can lead to data silos requiring system engineers to juggle multiple dashboards, and that might cause them to miss something crucial.
Agencies are better served with an observability strategy that provides a holistic, end-to-end view of network paths and devices that they otherwise have no control over. In doing so, they can literally look inside the cloud and automatically spot network slowdowns, determine the root cause of an outage, and accelerate issue resolution, whether it’s on-premises or in the cloud.
There is also a security issue with serverless architectures. The attack surface (which can include services from multiple vendors) is much greater than an agency could typically have with an in-house server. Agencies must assess the cloud services provider's risk management program, both internally and across third parties.
LEARN MORE: The State Department relies on scalable cloud architecture.
Calculate the Pros and Cons of Serverless Architecture
For agencies weighed down by operational maintenance and the imperative of IT modernization, serverless architecture could be a cost-effective solution for rapid application delivery. However, it isn’t a universal solution.
Agencies must do their due diligence and discover ways to enhance the power of this emerging trend while overcoming its disadvantages. They must ask their cloud provider about their security program, determine which applications are appropriate for serverless and figure out how to make these environments more observable before they sign on the dotted line.
