Fremer: Lightweight and Effective Frequency Transformer for Workload Forecasting in Cloud Services

2025-07-17

AI Summary

To address the high computational cost of Transformer models and their inefficiency in modeling complex periodic workload sequences in large-scale cloud environments, this paper proposes Fremer, a lightweight frequency-domain forecasting model. Methodologically, Fremer is the first to incorporate Fourier transform into workload prediction, explicitly capturing multi-scale periodicity; it introduces a cross-frequency interaction module and a lightweight attention mechanism to enable efficient global modeling directly in the frequency domain. Contributions include an optimal trade-off among accuracy, parameter count (reduced by 62%), and inference speed (accelerated 3.1×), alongside four high-quality, open-sourced workload datasets derived from ByteDance’s production systems. Experiments demonstrate that Fremer achieves average improvements of 5.5% in MSE, 4.7% in MAE, and 8.6% in SMAPE across multiple benchmarks; in Kubernetes autoscaling, it reduces latency by 18.78% and saves 2.35% in resource consumption.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Planning, Routing, and Scheduling: Planning with Language Models
• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Workload forecasting is pivotal in cloud service applications, such as auto-scaling and scheduling, with profound implications for operational efficiency. Although Transformer-based forecasting models have demonstrated remarkable success in general tasks, their computational efficiency often falls short of the stringent requirements in large-scale cloud environments. Given that most workload series exhibit complicated periodic patterns, addressing these challenges in the frequency domain offers substantial advantages. To this end, we propose Fremer, an efficient and effective deep forecasting model. Fremer fulfills three critical requirements: it demonstrates superior efficiency, outperforming most Transformer-based forecasting models; it achieves exceptional accuracy, surpassing all state-of-the-art (SOTA) models in workload forecasting; and it exhibits robust performance for multi-period series. Furthermore, we collect and open-source four high-quality, open-source workload datasets derived from ByteDance's cloud services, encompassing workload data from thousands of computing instances. Extensive experiments on both our proprietary datasets and public benchmarks demonstrate that Fremer consistently outperforms baseline models, achieving average improvements of 5.5% in MSE, 4.7% in MAE, and 8.6% in SMAPE over SOTA models, while simultaneously reducing parameter scale and computational costs. Additionally, in a proactive auto-scaling test based on Kubernetes, Fremer improves average latency by 18.78% and reduces resource consumption by 2.35%, underscoring its practical efficacy in real-world applications.

Problem

• Improve computational efficiency for cloud workload forecasting
• Enhance accuracy in predicting complex periodic workload patterns
• Reduce resource consumption in real-world cloud auto-scaling

Innovation

• Lightweight frequency transformer for cloud forecasting
• Superior efficiency and accuracy in workload prediction
• Robust performance for multi-period workload series