Metronome: Differentiated Delay Scheduling for Serverless Functions
Zhuangbin Chen, Juzheng Zheng, Zibin Zheng
TL;DR
The paper tackles the challenge of scheduling serverless functions by recognizing that data locality and infrastructure locality both influence performance, and that function execution times are highly heterogeneous. It introduces Metronome, a differentiated delay scheduling framework that uses an online Random Forest Regression model to predict per-node execution times and guide locality-aware decisions while preserving SLAs. Through profiling, online learning, and a low-overhead scheduling loop implemented on OpenLambda, Metronome achieves substantial reductions in mean and tail latency across data-intensive and dependency-heavy workloads, with strong scalability under concurrency. The work highlights that naive rule-based delays can harm SLA compliance and data locality alone is insufficient, proposing a practical, adaptive approach with broad applicability to serverless platforms.
Abstract
Function-as-a-Service (FaaS) computing is an emerging cloud computing paradigm for its ease-of-management and elasticity. However, optimizing scheduling for serverless functions remains challenging due to their dynamic and event-driven nature. While data locality has been proven effective in traditional cluster computing systems through delay scheduling, its application in serverless platforms remains largely unexplored. In this paper, we systematically evaluate existing delay scheduling methods in serverless environments and identify three key observations: 1) delay scheduling benefits vary significantly based on function input characteristics; 2) serverless computing exhibits more complex locality patterns than cluster computing systems, encompassing both data locality and infrastructure locality; and 3) heterogeneous function execution times make rule-based delay thresholds ineffective. Based on these insights, we propose Metronome, a differentiated delay scheduling framework that employs predictive mechanisms to identify optimal locality-aware nodes for individual functions. Metronome leverages an online Random Forest Regression model to forecast function execution times across various nodes, enabling informed delay decisions while preventing SLA violations. Our implementation on OpenLambda shows that Metronome significantly outperforms baselines, achieving 64.88%-95.83% reduction in mean execution time for functions, while maintaining performance advantages under increased concurrency levels and ensuring SLA compliance.