AARC: Automated Affinity-aware Resource Configuration for Serverless Workflows

TL;DR

The paper tackles inefficiencies in serverless workflows caused by coupled CPU–memory resource configurations. It introduces AARC, an affinity-aware resource configuration framework with two components: Graph-Centric Scheduler, which decomposes workflows into a weighted DAG and identifies the critical path, and Priority Configurator, which uses priority scheduling to tune CPU and memory allocations along that path and related sub-paths to meet end-to-end $SLO$. The work demonstrates that decoupling resources expands the configuration space but can be navigated efficiently through the proposed architecture, achieving around an 85% reduction in search runtime and about 50% cost savings compared with state-of-the-art baselines, while preserving $SLO$ compliance. These results suggest a practical path to automated, cost-effective resource provisioning for complex serverless workflows, with an additional input-aware extension that further optimizes configurations for varying input characteristics."

Abstract

Serverless computing is increasingly adopted for its ability to manage complex, event-driven workloads without the need for infrastructure provisioning. However, traditional resource allocation in serverless platforms couples CPU and memory, which may not be optimal for all functions. Existing decoupling approaches, while offering some flexibility, are not designed to handle the vast configuration space and complexity of serverless workflows. In this paper, we propose AARC, an innovative, automated framework that decouples CPU and memory resources to provide more flexible and efficient provisioning for serverless workloads. AARC is composed of two key components: Graph-Centric Scheduler, which identifies critical paths in workflows, and Priority Configurator, which applies priority scheduling techniques to optimize resource allocation. Our experimental evaluation demonstrates that AARC achieves substantial improvements over state-of-the-art methods, with total search time reductions of 85.8% and 89.6%, and cost savings of 49.6% and 61.7%, respectively, while maintaining SLO compliance.

Figures (8)

• Figure 1: Architecture of Three Workflows.
• Figure 2: Runtime and Cost with Decoupled Resources.
• Figure 3: Bayesian Optimization Search for Chatbot.
• Figure 4: System Architecture.
• Figure 5: Overall Sample Cost and Runtime Comparison.