Serverless computing represents one more step on the path away from cumbersome monolithic applications to more agile software approaches that speed development and deployment time. Such “cloud-native” applications are designed to leverage the flexible infrastructure of the cloud by abstracting application logic from the underlying server hardware. Cloud-native architectures include virtual machines, containers, and serverless computing.

Contrary to its name, serverless compute does indeed involve physical servers, usually provisioned in a cloud data center. The term “serverless” indicates that code is deployed on a server only as and when needed, without requiring a technician to allocate or manage the server capacity. Thus, with serverless computing, businesses can run snippets of code automatically, as needed, based on “triggers” coded into other applications or microservices. Compared with other architectures, the serverless model can be more cost- and labor-efficient, especially for dynamic applications.

In the past few years, all major cloud service providers have launched serverless offerings. The most common are event-driven serverless compute capabilities, called Functions-as-a-Service, that allow customers to deploy serverless applications into their clouds, with the provider managing the infrastructure capacity.

In this report, we explore the value of serverless architectures for enterprise developers and the associated growth opportunities for public cloud service providers. We also review serverless offerings of leading providers including AWS, Microsoft, Google, IBM, and Oracle.