Enhancing the performance of GPU acceleration in virtual environments: Thoroughly benchmarking the rigidity of mediated device passthrough

[EN] Virtualization has been the key element for the growth of cloud computing. Historically, GPUs have been complex devices to virtualize efficiently. The mediated passthrough mechanism is usually leveraged. However, it implies a rigid association between the virtual domain and the virtual GPU, whi...

Descripción completa

Detalles Bibliográficos
Autores: Prades, Javier|||0000-0003-3349-2200, Silla, Federico|||0000-0002-6435-1200, Reaño, Carlos
Tipo de recurso: artículo
Fecha de publicación:2026
País:España
Institución:Universitat Politècnica de València (UPV)
Repositorio:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Idioma:inglés
OAI Identifier:oai:riunet.upv.es:10251/228369
Acceso en línea:https://riunet.upv.es/handle/10251/228369
Access Level:acceso abierto
Palabra clave:CUDA
GPU acceleration
Virtualization
NVIDIA vGPU
NGS
RCUDA
Mediated device passthrough
Descripción
Sumario:[EN] Virtualization has been the key element for the growth of cloud computing. Historically, GPUs have been complex devices to virtualize efficiently. The mediated passthrough mechanism is usually leveraged. However, it implies a rigid association between the virtual domain and the virtual GPU, which impairs overall system GPU performance. In this paper we propose the use of Network GPGPU system (NGS) to improve the performance of GPUs in virtual domains. Our proposal is compared to NVIDIA vGPU, the most widely used mechanism for virtualizing CUDA-enabled GPUs today. Results show throughput benefits of approximately 20%, speedup of up to 28% in the execution time of the applications, improved overall GPU utilization (over 85%), and lower energy consumption per job (up to 15.34%).