Scholarlink

<p>To mitigate the increasingly common underutilization of computational resources in modern GPUs, spatial sharing methods enable multiple applications to use them simultaneously. This work presents a comprehensive evaluation of NVIDIA’s primary technologies to achieve that goal: multi-process service (MPS) and multi-instance GPU (MIG). Our findings reveal a crucial trade-off between MPS’s flexibility and MIG’s isolation, and provide many key insights for improving the co-execution strategy according to job profiles. In the most favorable scenarios, MPS improves performance by up to 30% and reduces energy by about 20%, using its provisioning option to avoid resource monopolization. However, under memory contention, it suffers severe degradation, worsening performance by around 30%. Conversely, MIG’s full hardware isolation resolves memory contention, leading to more consistent improvements, but these gains are tempered by higher overhead, and its rigid scheme can degrade performance in certain cases.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A comprehensive evaluation of spatial co-execution on GPUs using MPS and MIG technologies

  • 期刊论文

  • 发表日期：

    2026年2月17日

查看全文
  • Jorge Villarrubia,
  • Luis Costero,
  • Francisco D. Igual,
  • Katzalin Olcoz

摘要

To mitigate the increasingly common underutilization of computational resources in modern GPUs, spatial sharing methods enable multiple applications to use them simultaneously. This work presents a comprehensive evaluation of NVIDIA’s primary technologies to achieve that goal: multi-process service (MPS) and multi-instance GPU (MIG). Our findings reveal a crucial trade-off between MPS’s flexibility and MIG’s isolation, and provide many key insights for improving the co-execution strategy according to job profiles. In the most favorable scenarios, MPS improves performance by up to 30% and reduces energy by about 20%, using its provisioning option to avoid resource monopolization. However, under memory contention, it suffers severe degradation, worsening performance by around 30%. Conversely, MIG’s full hardware isolation resolves memory contention, leading to more consistent improvements, but these gains are tempered by higher overhead, and its rigid scheme can degrade performance in certain cases.