NVIDIA FLARE 跨 Docker、Kubernetes 和 Slurm 扩展联邦学习
原标题：Scaling Federated Learning Across Docker, Kubernetes, and Slurm with NVIDIA FLARE
AI 摘要
NVIDIA 发布技术博客，介绍 NVIDIA FLARE 如何通过两层架构将联邦学习中的持久化联邦服务与作业执行分离，使同一联邦中的不同站点可分别使用 Docker、Kubernetes 或 Slurm 作为执行后端。FLARE 2.8 已支持 Docker 和 Kubernetes 部署，2.9 新增 Slurm 支持。该设计让各站点保留对计算资源、数据集、镜像和调度策略的本地控制权，降低跨异构基础设施协作的门槛。
正文节选
Federated learning (FL) projects often begin with a straightforward setup: one server, a few clients, and one dataset at each site. As those projects grow, the challenge shifts from running an algorithm to operating shared infrastructure. GPUs must be allocated when jobs need them, multiple research studies must remain separated, and every participating organization must retain control of its own data, secrets, and compute policies. That operational complexity grows when organizations rely on di