Konflux: Optimized Function Fusion for Serverless Applications

2026-01-16

AI Summary

This work addresses the challenge of optimizing cost and latency in serverless applications, where the vast configuration space of function fusion renders exhaustive exploration impractical. To overcome this, the authors propose a localized evaluation method based on FaaS platform simulation that efficiently explores the entire fusion configuration space without requiring modifications to the deployed application. This approach enables, for the first time, scalable analysis of all possible fusion configurations for complex applications by combining systematic enumeration with multi-resource constraint modeling. The study reveals that optimal fusion strategies are highly sensitive to specific pricing models and occupy only a minuscule fraction of the configuration space. Experimental results demonstrate that the method substantially reduces optimization overhead and accurately identifies Pareto-optimal fusion configurations dictated by the underlying pricing model.

Technology Category

• Search and Optimization: Learning to Search
• Constraint Satisfaction and Optimization: Applications
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Function-as-a-Service (FaaS) has become a central paradigm in serverless cloud computing, yet optimizing FaaS deployments remains challenging. Using function fusion, multiple functions can be combined into a single deployment unit, which can be used to reduce cost and latency of complex serverless applications comprising multiple functions. Even in small-scale applications, the number of possible fusion configurations is vast, making brute-force benchmarking in production both cost- and time-prohibitive. In this paper, we present a system that can analyze every possible fusion setup of complex applications. By emulating the FaaS platform, our system enables local experimentation, eliminating the need to reconfigure the live platform and significantly reducing associated cost and time. We evaluate all fusion configurations across a number of example FaaS applications and resource limits. Our results reveal that, when analyzing cost and latency trade-offs, only a limited set of fusion configurations represent optimal solutions, which are strongly influenced by the specific pricing model in use.

Problem

• Function-as-a-Service
• function fusion
• serverless optimization
• cost-latency trade-off
• fusion configuration

Innovation

• function fusion
• serverless computing
• FaaS optimization
• cost-latency trade-off
• emulation-based analysis