发表机构
Wuhan University; Beijing University of Posts and Telecommunications（武汉大学; 北京邮电大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对云微服务调度中资源动态失衡与异构耦合问题，提出MCRL2，利用多资源交叉注意力表示学习增强强化学习，在真实集群轨迹上显著提升负载均衡、调度成功率并降低平均完成时间。
AI 中文摘要
高效的微服务调度对于维持数据中心各节点间的负载均衡和确保高质量服务至关重要。然而，在实际中实现这一目标仍具挑战性，原因在于波动工作负载下的动态资源失衡、跨多个资源维度的非线性耦合以及微服务资源需求的异构性。尽管基于强化学习的方法已展现出潜力，但它们难以捕捉异构资源间复杂的相互依赖关系，并忽视了学习信息丰富的系统表示的重要性。为解决这些局限，我们提出了MCRL2，一种新颖的强化学习方法，通过多资源交叉注意力机制增强表示学习以用于微服务调度。具体而言，我们首先提出MCRL，一种新颖的表示学习方法，通过多资源交叉注意力机制捕捉节点、资源和微服务之间结构化且信息丰富的交互。然后，MCRL2通过MCRL增强的演员-评论家架构结合最大熵目标来增强强化学习，提升系统状态表达能力，从而带来更稳定和有效的调度决策。在真实生产集群轨迹上的大量实验表明，MCRL2在负载均衡、调度成功率和平均完成时间方面，在不同工作负载模式下显著优于现有基线。
英文摘要
Efficient microservice scheduling is crucial for maintaining load balance across nodes in data centers and ensuring high quality of service. However, achieving this in practice remains challenging due to dynamic resource imbalance under fluctuating workloads, nonlinear coupling across multiple resource dimensions, and the heterogeneity of microservice resource demands. While reinforcement learning-based approaches have shown promise, they struggle to capture the complex interdependencies among heterogeneous resources and neglect the importance of learning informative system representations. To address these limitations, we propose MCRL2, a novel reinforcement learning approach augmented with multi-resource cross-attention-based representation learning for microservice scheduling. Specifically, we first propose MCRL, a novel representation learning approach that captures structured and informative interactions among nodes, resources, and microservices via a multi-resource cross-attention mechanism. Then, MCRL2 augments reinforcement learning through MCRL-enhanced actor-critic architecture combined with a maximum entropy objective, improving system state expressiveness and leading to more stable and effective scheduling decisions. Extensive experiments on real production cluster traces demonstrate that MCRL2 significantly outperforms existing baselines in load balancing, scheduling success rate and average completion time across diverse workload patterns.
Comments15 pages, 14 figures
