Hyper-V GPU passthrough

Overview

Hyper-V GPU passthrough에 대한 기술 문서입니다.

Summary

• 무엇인가? - Hyper-V GPU passthrough
• 왜 필요한가? - HPC 및 서버 환경에서 필수 개념
• 언제 사용하는가? - 서버 구성, 성능 튜닝, 문제 해결 시

Purpose

이 문서가 존재하는 이유

• Goal: Hyper-V GPU passthrough에 대한 기술 정보 제공
• Scope: Hyper-V GPU passthrough의 개념, 사용법, 설정
• Non-goals: 다른 주제로의 확장

Key Concepts

Concept	Description	Related
Hyper-V GPU passthrough	HPC/서버 환경에서 중요한 기술 개념	Linux, Server

Detailed Explanation

GPU partitioning is a method to partition off resources from your graphics card so that it can be used inside your VMs. Hyper-V supports is GPU passthrough, a feature that allows virtual machines to access the host’s GPU directly. this feature allows graphics intensive workloads like gaming or video editing on Hyper-V guest. This posting explains how to set GPU to Hyper-V window VM Test to see if your GPU can be partitioned at all. On the Host, open a Powershell prompt as administrator. Get-VMPartitionableGpu (win10) or Get-VMHostPartitionableGpu (win11)

• Need a GPU that supports GPU virtualization, such as NVIDIA GRID or AMD MxGPU technology
• Your CPU needs to support IOMMU (Input-Output Memory Management Unit) virtualization, which is required for GPU passthrough
• GPU passthrough is not supported for virtual machines that use Dynamic Memory
• GPU passthrough is only available on specific editions of Windows Server, such as the Datacenter edition
• brokeDude2901 introduced how to enable GPU passthrough to Ubuntu VM using dxgkrnl_ubuntu kernel on github

Best Practices

• 최신 버전 사용 권장
• 공식 문서 참고
• 테스트 환경에서 먼저 검증

References

• Hyper-V GPU passthrough

Related Pages

• Linux
• Server
• Hardware
• Network

Knowledge Graph

Related

→ CUDA → GPU → NVIDIA GPU → NVLink → AMD GPUs → MIG → NVIDIA GPUs → Tensor Core