Guardian: Safe GPU Sharing in Multi-Tenant Environments
Manos Pavlidakis, Giorgos Vasiliadis, Stelios Mavridis, Anargyros Argyros, Antony Chazapis, Angelos Bilas · 2024
Modern GPU applications, such as machine learning (ML), can only partially utilize GPUs, leading to GPU underutilization in cloud environments. Sharing GPUs across multiple applications from different tenants can improve resource utilization and consequently cost, energy, and power efficiency. However, GPU sharing creates memory safety concerns because kernels must share a single GPU address space. Existing spatial-sharing mechanisms either lack fault isolation for memory accesses or require static partitioning, which leads to limited deployability or low utilization.