A computer and a laptop with clouds in the background.

Serverless computing has become a popular model for building and deploying applications in the cloud. It offers many benefits, including scalability, cost-effectiveness, and simplified deployment. However, there are certain scenarios where serverless computing may not be the best choice. In this blog post, we will discuss some scenarios where serverless computing may not be the optimal solution.

Basic Architecture of a Serverless Application

Long-running processes

Serverless computing is designed for short-lived, event-driven processes. For processes that run for a longer duration, such as batch processing or machine learning training jobs, the cost of running these processes on a serverless platform can quickly become expensive. In these cases, a traditional compute platform or container orchestration solution may be more suitable.

High-performance computing

Serverless computing is not well suited for high-performance computing (HPC) workloads that require low-latency, high-bandwidth networking, and high-performance hardware. These workloads often require specialized hardware or software configurations that are not available on serverless platforms.

Resource-intensive workloads

Serverless computing platforms often have resource limits on the amount of CPU, memory, and disk space that can be allocated to a function. For resource-intensive workloads such as data processing, image or video rendering, or scientific simulations, these limits can be a hindrance. In these cases, a dedicated compute platform may be a better fit.

Stateful applications

Serverless computing platforms are designed for stateless applications that can be easily scaled horizontally. For stateful applications, such as databases or applications that require persistent connections, serverless computing may not be the best choice. These types of applications often require specialized configurations and may be better suited to a traditional compute platform.

Custom runtime requirements

Serverless platforms typically support a limited set of programming languages and runtimes. If your application requires a custom runtime or language that is not supported by the serverless platform, you may need to choose a different compute platform.


In conclusion, while serverless computing offers many benefits, it is not always the best choice for every scenario. For long-running processes, high-performance computing, resource-intensive workloads, stateful applications, and custom runtime requirements, a traditional compute platform may be a better fit. It is important to carefully consider the requirements of your application and choose the most appropriate platform for your needs.

To the Cloud, and Beyond…

Join thousands of industry experts and get the latest tips and tricks from the cloud industry.