Multi-Level Service Performance Forecasting via Spatiotemporal Graph Neural Networks

2025-08-09

AI Summary

Accurately predicting performance fluctuations in distributed backend systems is challenging due to complex, multi-tier service call hierarchies. Method: This paper proposes a spatiotemporal graph neural network framework that unifies service topology into a graph structure, incorporates runtime features and position-aware temporal encodings, and jointly leverages graph convolutional networks (to capture cross-service higher-order dependencies) and gated recurrent units (to model temporal dynamics), enabling end-to-end optimization over deeply nested architectures. Contribution/Results: Evaluated on a large-scale public cluster dataset, the model achieves significant improvements over state-of-the-art methods in MAE, RMSE, and R². It demonstrates strong robustness across varying load intensities and topology complexities. The framework establishes a scalable, high-accuracy, general-purpose paradigm for microservice performance prediction.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

This paper proposes a spatiotemporal graph neural network-based performance prediction algorithm to address the challenge of forecasting performance fluctuations in distributed backend systems with multi-level service call structures. The method abstracts system states at different time slices into a sequence of graph structures. It integrates the runtime features of service nodes with the invocation relationships among services to construct a unified spatiotemporal modeling framework. The model first applies a graph convolutional network to extract high-order dependency information from the service topology. Then it uses a gated recurrent network to capture the dynamic evolution of performance metrics over time. A time encoding mechanism is also introduced to enhance the model's ability to represent non-stationary temporal sequences. The architecture is trained in an end-to-end manner, optimizing the multi-layer nested structure to achieve high-precision regression of future service performance metrics. To validate the effectiveness of the proposed method, a large-scale public cluster dataset is used. A series of multi-dimensional experiments are designed, including variations in time windows and concurrent load levels. These experiments comprehensively evaluate the model's predictive performance and stability. The experimental results show that the proposed model outperforms existing representative methods across key metrics such as MAE, RMSE, and R2. It maintains strong robustness under varying load intensities and structural complexities. These results demonstrate the model's practical potential for backend service performance management tasks.

Problem

• Forecasting performance fluctuations in distributed backend systems
• Modeling multi-level service call structures with spatiotemporal dependencies
• Achieving high-precision regression of future service performance metrics

Innovation

• Spatiotemporal graph neural networks for forecasting
• Graph convolutional and gated recurrent networks integration
• Time encoding mechanism for non-stationary sequences