AAPA: An Archetype-Aware Predictive Autoscaler with Uncertainty Quantification for Serverless Workloads on Kubernetes

TL;DR

AAPA introduces an archetype-aware autoscaler for Kubernetes serverless workloads that classifies workload patterns and applies tailored scaling with uncertainty quantification, improving performance and reliability.

Contribution

The paper proposes a novel workload classification-based autoscaling method with uncertainty modeling and releases a large dataset for evaluation.

Findings

Reduces SLO violations by up to 50%

Lowers latency by 40% compared to HPA

Increases resource usage 2-8x under spike conditions

Abstract

Serverless platforms such as Kubernetes are increasingly adopted in high-performance computing, yet autoscaling remains challenging under highly dynamic and heterogeneous workloads. Existing approaches often rely on uniform reactive policies or unconditioned predictive models, ignoring both workload semantics and prediction uncertainty. We present AAPA, an archetype-aware predictive autoscaler that classifies workloads into four behavioral patterns -- SPIKE, PERIODIC, RAMP, and STATIONARY -- and applies tailored scaling strategies with confidence-based adjustments. To support reproducible evaluation, we release AAPAset, a weakly labeled dataset of 300,000 Azure Functions workload windows spanning diverse patterns. AAPA reduces SLO violations by up to 50% and lowers latency by 40% compared to Kubernetes HPA, albeit at 2-8x higher resource usage under spike-dominated conditions. To assess trade-offs, we propose the Resource Efficiency Index (REI), a unified metric balancing performance, cost, and scaling smoothness. Our results demonstrate the importance of modeling workload heterogeneity and uncertainty in autoscaling design.