HRS: Hybrid Representation Framework with Scheduling Awareness for Time Series Forecasting in Crowdsourced Cloud-Edge Platforms

2025-08-18

AI Summary

To address the degradation of QoS and SLA violations caused by highly time-varying and bursty streaming workloads in crowdsourced cloud–edge platforms (CCPs), this paper proposes a scheduling-aware multimodal load forecasting framework. The method innovatively fuses numerical time-series data with image-based representations to model extreme workload dynamics; introduces a scheduling-aware loss function that captures the asymmetric impact of prediction errors on resource scheduling decisions; and jointly optimizes load forecasting and scheduling policies. Extensive experiments across four real-world datasets demonstrate that, compared to ten state-of-the-art baselines, the proposed framework reduces SLA violation rate by 63.1% and total profit loss by 32.3%, while significantly improving resource utilization efficiency and service reliability.

Technology Category

• Planning, Routing, and Scheduling: Scheduling under Uncertainty
• Machine Learning: Time-Series/Data Streams
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Systems and Infrastructure for Web, Mobile and WoT: Cloud, edge and content delivery systems for the Web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

With the rapid proliferation of streaming services, network load exhibits highly time-varying and bursty behavior, posing serious challenges for maintaining Quality of Service (QoS) in Crowdsourced Cloud-Edge Platforms (CCPs). While CCPs leverage Predict-then-Schedule architecture to improve QoS and profitability, accurate load forecasting remains challenging under traffic surges. Existing methods either minimize mean absolute error, resulting in underprovisioning and potential Service Level Agreement (SLA) violations during peak periods, or adopt conservative overprovisioning strategies, which mitigate SLA risks at the expense of increased resource expenditure. To address this dilemma, we propose HRS, a hybrid representation framework with scheduling awareness that integrates numerical and image-based representations to better capture extreme load dynamics. We further introduce a Scheduling-Aware Loss (SAL) that captures the asymmetric impact of prediction errors, guiding predictions that better support scheduling decisions. Extensive experiments on four real-world datasets demonstrate that HRS consistently outperforms ten baselines and achieves state-of-the-art performance, reducing SLA violation rates by 63.1% and total profit loss by 32.3%.

Problem

• Accurate load forecasting under traffic surges in CCPs
• Balancing SLA compliance and resource expenditure
• Improving scheduling decisions with hybrid representations

Innovation

• Hybrid representation framework for load dynamics
• Scheduling-Aware Loss for asymmetric errors
• Combines numerical and image-based representations