search

4paradigm / k8s-vgpu-scheduler

OpenAIOS vGPU device plugin for Kubernetes is originated from the OpenAIOS project to virtualize GPU device memory, in order to allow applications to access larger memory space than its physical capacity. It is designed for ease of use of extended device memory for AI workloads.
Apache License 2.0
489 stars 93 forks source link

Scheduling adds three scoring dimensions. #39

Open zrbilvoe opened 3 months ago

zrbilvoe commented 3 months ago

为了使调度更加平衡, 增加了宿主层面3个维度的打分机制, 通过启动参数 total_core_node_weight, count_node_weight, node_podantiaffinity_weight 控制不同维度的打分权重

  1. total_core_node_weight 宿主上算力切分之后的剩余算力, 剩余算力越多得分越高
  2. count_node_weight 宿主上可部署的pod总数, 可部署的pod数量越多得分越高
  3. node_podantiaffinity_weight 同一deploy应用在同一宿主的反亲和性, 属于同一deploy的pod在同一宿主上数量越少得分越高, 使得同一deploy应用部署更加分散