一句话：提出面向多 GPU 集合通信的领域特定语言 MSCCLang，用于表达针对网络拓扑与应用模式定制的集合算法。
随着机器学习模型参数规模达到百万乃至十亿级，其训练与服务日益依赖大规模多 GPU 系统。模型变大且使用的 GPU 增多后，集合通信成为性能瓶颈。针对特定网络拓扑和应用通信模式定制的集合算法可以缓解该瓶颈并帮助应用扩展，但正确且高效地实现这类算法十分困难。本文提出 MSCCLang（Microsoft Collective Communication Language），一种面向集合通信的语言，旨在为开发者提供表达定制集合算法的抽象。给出的摘要仅说明动机与语言名称，未提供具体语言设计、编译或运行时支持以及实验评估结果。
Machine learning models with millions or billions of parameters are increasingly trained and served on large multi-GPU systems. As models grow in size and execute on more GPUs, collective communication becomes a bottleneck. Custom collective algorithms optimized for both particular network topologies and application-specific communication patterns can alleviate this bottleneck and help these applications scale. However, implementing correct and efficient custom algorithms is challenging.