一句话：提出 Imbres 微服务系统，利用负载不均衡动态迁移负载并协同扩缩容，降低资源与 SLA 违规。
在共享集群中，同一微服务的容器常出现时空性能不均衡，传统负载均衡为满足 SLA 往往过度配置资源。本文反向利用不均衡，提出 Imbres 微服务系统，将负载迁移、连接管理与资源扩缩容协同优化。其核心是仅依赖在线延迟梯度进行快速响应，无需离线剖析，并支持上下游容器间部分连接约束。基于真实微服务基准的评估表明，相较当前先进系统，Imbres 最多减少 62% 资源分配，并将 SLA 违规概率最多降低 82%。
In a unified resource scheduling architecture, containers within the same microservice often encounter temporal and spatial performance imbalance when deployed in large-scale shared clusters. As a result, the commonly employed load-balancing approach often leads to substantial resource wastage as applications are frequently over-provisioned to meet service level agreements (SLAs). In this paper, we utilize an alternative approach by leveraging load imbalance. The central concept involves the dynamic load shifting across microservice containers with a focus on imbalance awareness. However, achieving seamless integration between load shifting and resource scaling, while accommodating the demands of partial connection between upstream and downstream containers, remains a challenge. To address this challenge, we introduce Imbres-a new microservice system that optimizes load shifting, connection management, and resource scaling in tandem. One significant advantage of Imbres lies in its rapid responsiveness, relying solely on online gradients of latency, eliminating the need for offline profiling. Evaluation using real microservice benchmarks reveals that Imbres reduces resource allocation by up to 62% and decreases SLA violation probability by up to 82%, compared to state-of-the-art systems.
@inproceedings{LuoLLXZ025,
title = {Embracing Imbalance: Dynamic Load Shifting among Microservice Containers in Shared Clusters},
author = {Shutian Luo and Jianxiong Liao and Chenyu Lin and Huanle Xu and Zhi Zhou and Chengzhong Xu},
booktitle = {ASPLOS},
year = {2025},
doi = {10.1145/3676641.3716255}
}