Leveraging Core and Uncore Frequency Scaling for Power-Efficient Serverless Workflows

2024-07-25

arXiv.org

AI Summary

End-to-end QoS and power co-optimization for serverless workflows remains challenging due to the lack of holistic hardware-aware resource control. Method: This paper proposes Ωkypous, the first QoS-aware power management framework for serverless that jointly orchestrates dynamic voltage and frequency scaling (DVFS) across both CPU cores and uncore components (e.g., memory controllers, interconnects). It introduces core–uncore frequency co-scaling as an intrinsic workflow optimization dimension and builds a gray-box model to accurately predict function latency and power consumption under arbitrary multi-frequency configurations. The framework then solves for the optimal frequency assignment subject to strict QoS (SLO) constraints. Results: Evaluated on the Azure Trace dataset, Ωkypous reduces system power consumption by 20% over state-of-the-art methods while significantly lowering SLO violation rates—demonstrating both the effectiveness and necessity of joint core–uncore DVFS for energy-efficient, QoS-guaranteed serverless execution.

Technology Category

• Machine Learning: Hardware-aware ML
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Planning, Routing, and Scheduling: Plan Execution and Monitoring

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Serverless workflows have emerged in FaaS platforms to represent the operational structure of traditional applications. With latency propagation effects becoming increasingly prominent, step-wise resource tuning is required to address the end-to-end Quality-of-Service (QoS) requirements. Modern processors' allowance for fine-grained Dynamic Voltage and Frequency Scaling (DVFS), coupled with the intermittent nature of serverless workflows presents a unique opportunity to reduce power while meeting QoS. In this paper, we introduce a QoS-aware DVFS framework for serverless workflows. {Omega}kypous regulates the end-to-end latency of serverless workflows by supplying the system with the Core/Uncore frequency combination that minimizes power consumption. With Uncore DVFS enriching the efficient power configurations space, we devise a grey-box model that accurately projects functions' execution latency and power, to the applied Core and Uncore frequency combination. To the best of our knowledge, {Omega}kypous is the first work that leverages Core and Uncore DVFS as an integral part of serverless workflows. Our evaluation on the analyzed Azure Trace, against state-of-the-art (SotA) power managers, demonstrates a power consumption reduction of 20% while minimizing QoS violations.

Problem

• Optimize power efficiency in serverless workflows while meeting latency SLOs
• Dynamically adjust Core and Uncore frequencies using DVFS for energy savings
• Reduce power consumption without violating end-to-end latency constraints

Innovation

• SLO-driven DVFS framework for serverless workflows
• Grey-box model predicts latency and power
• Closed-loop control adjusts Core and Uncore frequencies