The rise of the serverless cloud

Paradoxically, both “serverless cloud” and “serverless computing” are misnomers, as neither eliminate the need for servers.
Paul Stuttard
By Paul Stuttard, Director, Duxbury Networking.
Johannesburg, 09 Nov 2022

The concept of the serverless cloud, first proposed in 2019, is gaining popularity due to increasing support from industry heavyweights such as Amazon, Microsoft and IBM.

Born out of the need to optimise resource usage and costs, the serverless cloud describes a concept that permits developers to build and run applications in the cloud without the need to manage servers.

Sometimes labelled a hosting and developer acceleration platform, it permits automatic scaling designed to increase agility and minimise costs, particularly as they relate to capacity planning, infrastructure management and maintenance.

The serverless cloud concept also allows developers and operators more time to focus on application logic and user requirements. It also gives them opportunities to boost their own productivity through the rapid delivery of new applications and features.

This is underlined by Gartner, the respected research and consulting firm, which says it represents a “next-generation technology able to redefine the way enterprises build, consume and integrate cloud-native applications”.

This brings us to the confusion that exists around the terms “serverless cloud” and its close cousin “serverless computing”. Paradoxically, both are misnomers. Neither eliminate the need for servers.

The serverless cloud and serverless computing do, however, eliminate the need for infrastructure set-up, configuration, provisioning and management. This is because the servers are abstracted away from application development and owned by service companies that run code in response to events. These companies automatically manage the allocation of machine resources required by that code.

This is in contrast with traditional cloud computing, in which end-users conventionally provision, manage and operate their own servers.

The serverless cloud and serverless computing eliminate the need for infrastructure set-up, configuration, provisioning and management.

Adding to the confusion, the serverless cloud and serverless computing are often viewed as performing comparable functions. While similarities do exist, the serverless cloud is, in essence, a concept, while serverless computing refers to an execution model through which cloud service providers dynamically manage the allocation of server resources.

More specifically, when it comes to managing this allocation, costs are often solely dependent on the time and memory allocated, which are generally on an “on-demand” basis.

In this light, serverless computing is seen as a compelling paradigm for the cost-effective deployment of applications and services.

Focusing on the serverless cloud, there are two main service offerings associated with it. These are function-as-a-service (FaaS) and backend-as-a-service (BaaS).

FaaS is the most widely used as it provides developers with a platform to easily develop and deploy code without any concern for infrastructure configurations. FaaS supports the idea of a pay-as-you-go service in which the infrastructure increases discreetly as an application's needs change.

Some industry researchers suggest FaaS and the term “serverless” could be used interchangeably. They say the FaaS platform automatically configures and maintains the execution context of functions and connects them to cloud services without requiring server provisioning by developers.

In similar vein, the BaaS model enables developers to manage only front-end functionality, while outsourcing backend functionality to third-party services.

Recently, the serverless cloud has been credited with incentivising cloud providers to innovate and invest in efficiency at every level. With this in mind, one of the major advantages of the concept is its ability to promote multi-cloud environments.

The multi-cloud approach is among the top 10 technology trends predicted for 2023. It is an approach in which organisations distribute applications and services over several clouds using a number of cloud types, including public, private and edge clouds.

Through a multi-cloud deployment, organisations are able to use multiple cloud providers and multiple platforms for different tasks. Organisations can therefore take advantage of distinctive blends of service and differently structured service-level agreements, as well as tailored levels of availability, security architectures, pricing, tools and resources to meet changing requirements.

Moreover, a multi-cloud deployment allows companies that do not wish to be tied to a single cloud provider for strategic reasons to explore opportunities to use best-of-breed services.

For example, a combination of cloud services could leverage multiple infrastructure-as-a-service vendors. Or different vendors for platform-as-a-service and software-as-a-service options could be employed.

By increasing the number of cloud vendors, security can be improved through failover options, which will allow operations to continue should the primary service provider suffer an outage or be security-compromised. In addition, a multi-cloud strategy facilitates workload portability delivered by modern, cloud-native platforms.

More significantly, a multi-cloud strategy will assist organisations in their digital transformation. Not only does it contribute towards an increase in an organisation’s computing power, it also permits the creation and testing of new products and services through access to, and the deployment of, many advanced analytical tools and services.

Against this backdrop, the IT industry can expect to see significant growth in specific serverless computing applications in 2023, along with the emergence of more general-purpose, serverless abstractions, which could host a variety of new software ecosystems catering to a wide range of specialised needs.