Characterizing FaaS Workflows on Public Clouds: The Good, the Bad and the Ugly

2025-09-26

AI Summary

Existing FaaS workflow platforms lack systematic empirical evaluation; their scaling behavior, inter-function data exchange, and billing models exhibit non-intuitive characteristics heavily influenced by platform design, dataflow patterns, and workload properties. Method: We conduct the first large-scale empirical study of AWS Step Functions, Azure Durable Functions, and Logic Apps, evaluating 25 microbenchmarks and real-world workflows across 132,000 invocations. We quantify performance using multidimensional metrics—including end-to-end latency, cold-start overhead, resource utilization, and cost. Contribution/Results: Our analysis reveals substantial disparities across platforms in performance, scalability, and cost-efficiency, exposing significant gaps between developer expectations and observed behavior. The findings provide empirically grounded guidance for configuration optimization and identify key research directions to enhance transparency, predictability, and efficiency of FaaS workflow systems.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Hardware-aware ML
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Function-as-a-service (FaaS) is a popular serverless computing paradigm for developing event-driven functions that elastically scale on public clouds. FaaS workflows, such as AWS Step Functions and Azure Durable Functions, are composed from FaaS functions, like AWS Lambda and Azure Functions, to build practical applications. But, the complex interactions between functions in the workflow and the limited visibility into the internals of proprietary FaaS platforms are major impediments to gaining a deeper understanding of FaaS workflow platforms. While several works characterize FaaS platforms to derive such insights, there is a lack of a principled and rigorous study for FaaS workflow platforms, which have unique scaling, performance and costing behavior influenced by the platform design, dataflow and workloads. In this article, we perform extensive evaluations of three popular FaaS workflow platforms from AWS and Azure, running 25 micro-benchmark and application workflows over 132k invocations. Our detailed analysis confirms some conventional wisdom but also uncovers unique insights on the function execution, workflow orchestration, inter-function interactions, cold-start scaling and monetary costs. Our observations help developers better configure and program these platforms, set performance and scalability expectations, and identify research gaps on enhancing the platforms.

Problem

• Characterizing complex interactions in FaaS workflow platforms
• Analyzing scaling and performance behavior of serverless workflows
• Investigating cost implications of FaaS workflow platform designs

Innovation

• Evaluated AWS and Azure FaaS workflow platforms
• Analyzed function execution and orchestration behaviors
• Identified performance scaling and cost optimization insights