LogoTensorFusion 文档
LogoTensorFusion 文档
首页文档

快速开始

TensorFusion 概览在 Kubernetes 安装在虚拟机/服务器安装(K3S)Helm 快速安装在宿主机/虚拟机安装TensorFusion 架构

应用操作

创建 AI 应用注解使用场景配置自动扩缩容迁移现有应用最佳实践

自定义AI基础设施

生产级部署指南QoS 级别与计价云厂商集成(BYOC)管理许可证

维护与优化

组件更新配置告警GPU 热迁移预加载模型优化 GPU 效率

故障排除

问题处理手册链路追踪/性能分析查询指标和日志

参考

对比

与 NVIDIA vGPU 对比与 MIG/MPS 对比与趋动科技对比与 NVIDIA Run:ai 对比与 HAMi 的对比

GPU

GPU is the Schema for the gpus API.

GPU is the Schema for the gpus API.

Resource Information

FieldValue
API Versiontensor-fusion.ai/v1
KindGPU
ScopeCluster

Status

GPUStatus defines the observed state of GPU.

PropertyTypeDescription
available *object
capacity *object
gpuModel *string
message *string
nodeSelector *objectThe host match selector to schedule worker pods
phase *string(default: Pending) Allowed values: Pending, Provisioning, Running, Unknown, Destroying, Migrating
runningAppsarray
usedBystringGPU is used by tensor-fusion or nvidia-operator This is the key to be compatible with nvidia-device-plugin to avoid resource overlap Hypervisor will watch kubelet device plugin to report all GPUs already used by nvidia-device-plugin GPUs will be grouped by usedBy to be used by different Pods, tensor-fusion annotation or nvidia-device-plugin resource block Allowed values: tensor-fusion, nvidia-device-plugin
uuid *string

目录

Resource Information
Status