The growing adoption of video-based AI models has created a pressing demand for high throughput, low latency inference systems. However, existing preprocessing frameworks—whether CPU or GPU based—struggle to keep up with the computational burdens of video decoding and data augmentation, resulting in suboptimal GPU utilization and degraded inference system performance.

Video-based deep learning (VDL) is increasingly used across diverse applications and has become highly popular, but it faces significant challenges in preprocessing highly compressed video data. Preprocessing pipelines are complex, requiring extensive engineering effort, and introduce computational bottlenecks, with latency exceeding GPU training time. Existing solutions partially mitigate these issues but remain inefficient and resource-constrained.

In cloud environments, GPU-based deep neural network (DNN) inference servers are required to meet the Service Level Objective (SLO) latency for each workload under a specified request rate, while also minimizing GPU resource consumption. However, previous studies have not fully achieved this objective. In this paper, we propose ParvaGPU, a technology that facilitates spatial GPU sharing for large-scale DNN inference in cloud computing. ParvaGPU integrates NVIDIA’s Multi-Instance GPU (MIG) and Multi-Process Service (MPS) technologies to enhance GPU utilization, with the goal of meeting the diverse SLOs of each workload and reducing overall GPU usage. Specifically, ParvaGPU addresses the challenges of minimizing underutilization within allocated GPU space partitions and external fragmentation in combined MIG and MPS environments. We conducted our assessment on multiple A100 GPUs, evaluating 11 diverse DNN workloads with varying SLOs. Our evaluation revealed no SLO violations and a significant reduction in GPU usage compared to state-of-the-art frameworks.