PulseAugur
EN
LIVE 21:22:55

New Graph Transformer models improve microservice tail latency prediction

By PulseAugur Editorial · [3 sources] ·

Two new research papers propose advanced methods for predicting tail latency in microservice systems. The first, STLGT, uses a graph transformer to model service dependencies and a temporal module for workload dynamics, showing improved accuracy and speed over existing methods. The second, USRFNet, employs a dual-stream learning approach to separate traffic and resource metrics, incorporating a gradient modulation strategy to address training imbalances and achieving significant reductions in prediction error. AI

IMPACT These new models offer improved accuracy and efficiency for predicting microservice tail latency, aiding proactive SLO management and system reliability.

RANK_REASON Two academic papers published on arXiv present novel methods for tail latency prediction in microservices.

Read on arXiv cs.LG →

AI-generated summary · Google Gemini · from 3 sources. How we write summaries →

New Graph Transformer models improve microservice tail latency prediction

COVERAGE [3]

  1. arXiv cs.AI TIER_1 English(EN) · Yongliang Ding, Qigong Bi, Peng Pu ·

    STLGT: A Scalable Trace-Based Linear Graph Transformer for Tail Latency Prediction in Microservices

    arXiv:2604.26422v1 Announce Type: cross Abstract: Accurate end-to-end tail-latency forecasting is critical for proactive SLO management in microservice systems. However, modeling long-range dependency propagation and non-stationary, bursty workloads while maintaining inference ef…

  2. arXiv cs.AI TIER_1 English(EN) · Peng Pu ·

    STLGT: A Scalable Trace-Based Linear Graph Transformer for Tail Latency Prediction in Microservices

    Accurate end-to-end tail-latency forecasting is critical for proactive SLO management in microservice systems. However, modeling long-range dependency propagation and non-stationary, bursty workloads while maintaining inference efficiency at scale remains challenging. We present …

  3. arXiv cs.LG TIER_1 English(EN) · Wenzhuo Qian, Hailiang Zhao, Jiayi Chen, Ziqi Wang, Tianlv Chen, Zhiwei Ling, Xinkui Zhao, Kingsum Chow, Albert Y. Zomaya, Shuiguang Deng ·

    Reliable Microservice Tail Latency Prediction via Decoupled Dual-Stream Learning and Gradient Modulation

    arXiv:2508.01635v2 Announce Type: replace Abstract: Microservice architectures enable scalable cloud-native applications; however, the distributed nature of these systems complicates the maintenance of strict Service Level Objectives. Accurately predicting window-level P95 tail l…

RELATED ENTITIES

RELATED TOPICS