🤖 AI Summary
This study addresses the opacity and complexity of pricing models in mainstream serverless Function-as-a-Service (FaaS) platforms, which hinder users from making cost-efficient deployment decisions. The authors systematically analyze the features and billing mechanisms of AWS Lambda, Azure Functions, and Google Cloud Functions, and present the first empirical cost evaluation by deploying representative workloads across multiple geographic regions. Their findings reveal that AWS consistently offers the lowest execution costs, while Azure incurs the highest, with significant price variations observed even within the same provider across different regions. By establishing a cross-platform and cross-regional cost benchmark, this work not only fills a critical gap in empirical FaaS cost analysis but also highlights substantial opportunities for optimizing current cloud pricing strategies.
📝 Abstract
Serverless Functions-as-a-Service providers have grown in their offering since inception a decade ago, with a myriad of new functionalities offered to end-users. These new features have also brought new, varied and at times complex pricing models that differ between providers. In this paper, we present Priceless, a detailed examination of the current state of the art of FaaS features, and their pricing models. We then perform a comparative price analysis of running example workloads across AWS Lambda, Microsoft Azure Functions and Google Cloud Functions. Our work finds significant cost differences both cross-provider, but also, cross-region within the provider. We find that AWS is the cheapest overall to run functions on, with Microsoft Azure being the most expensive for equivalent workloads.