NVIDIA Topograph：拓扑感知的 GPU 工作负载调度
Topology-Aware Workload Scheduling with NVIDIA Topograph
NVIDIA 推出 Topograph，用于拓扑感知的 GPU 工作负载调度。AI 工厂是功率受限系统，工作负载放置不当会割裂拓扑域、迫使流量走共享链路，降低吞吐、抬高作业成本，并让 GPU 在等待数据时白白消耗已分配电力。
AI factories are power-limited systems that deliver maximum value when fully optimized. GPU workload placement is a key optimization. Poor workload placement...
AI factories are power-limited systems that deliver maximum value when fully optimized. GPU workload placement is a key optimization. Poor workload placement fragments topology domains and forces traffic across shared links, reducing throughput, raising job costs, and leaving GPUs consuming provisioned power while waiting on data without advancing the workload. GPUs exchange data continuously…
来源：NVIDIA Developer Blog · developer.nvidia.com