一句话：ShapleyIQ 用因果图上的 Shapley 值量化微服务异常对端到端性能的影响，并以分裂不变性分解降低复杂度。
运维大规模微服务的经验表明，即使组件异常不可避免，仍需量化各组件对端到端性能指标的影响。在表示组件间复杂依赖的因果图上，分散检测到的异常即使相似，也可能对应不同含义与补救措施。为此，作者设计 ShapleyIQ：一种在线监控与诊断服务，旨在有效提升系统稳定性。该方法以对因果图上 Shapley 值的严格分析为指导。针对通用 Shapley 值计算复杂度随组件数指数增长的问题，论文提出一种新的分裂不变性（splitting invariance）性质，通过分解来降低计算复杂度。摘要未报告具体实验设置与结果，但将该服务定位为可提升大规模微服务系统稳定性的在线诊断工具。
Years of experience in operating large-scale microservice systems strengthens our belief that their individual components, with inevitable anomalies, still demand a quantification of the influences on the end-to-end performance indicators. On a causal graph that represents the complex dependencies between the system components, the scatteredly detected anomalies, even when they look similar, could have different implications with contrastive remedial actions. To this end, we design ShapleyIQ, an online monitoring and diagnosis service that can effectively improve the system stability. It is guided by rigorous analysis on Shapley values for a causal graph. Notably, a new property on splitting invariance addresses the challenging exponential computation complexity problem of generic Shapley values by decomposition.
@inproceedings{paperbot10111,
title = {ShapleyIQ: Influence Quantification by Shapley Values for Performance Debugging of Microservices},
author = {Ye Li and Jian Tan and Bin Wu and Xiao He and Feifei Li},
booktitle = {ASPLOS},
year = {2023},
doi = {10.1145/3623278.3624771}
}