Multi-model Machine Learning Inference Serving with GPU Spatial Partitioning
Seungbeom Choi, Sunho Lee, Yeonjae Kim et al. · arXiv (Cornell University) · 2021
As machine learning techniques are applied to a widening range of applications, high throughput machine learning (ML) inference servers have become critical for online service applications. Such ML inference servers pose two challenges: first, they must provide a bounded latency for each request to support consistent service-level objective (SLO), and second, they can serve multiple heterogeneous ML models in a system as certain tasks involve invocation of multiple models and consolidating multiple models can improve system utilization. To address the two requirements of ML inference servers, this paper proposes a new ML inference scheduling framework for multi-model ML inference servers. The paper first shows that with SLO constraints, current GPUs are not fully utilized for ML inference tasks. To maximize the resource efficiency of inference servers, a key mechanism proposed in this paper is to exploit hardware support for spatial partitioning of GPU resources. With the partitioning mechanism, a new abstraction layer of GPU resources is created with configurable GPU resources. The scheduler assigns requests to virtual GPUs, called gpu-lets, with the most effective amount of resources. The paper also investigates a remedy for potential interference effects when two ML tasks are running concurrently in a GPU. Our prototype implementation proves that spatial partitioning enhances throughput by 102.6% on average while satisfying SLOs.
Cited by 9 works. doi:10.48550/arxiv.2109.01611 Open-access PDF
Authors
References
- TensorFlow: A system for large-scale machine learning
- In-Datacenter Performance Analysis of a Tensor Processing Unit
- In-Datacenter Performance Analysis of a Tensor Processing Unit
- TVM: An Automated End-to-End Optimizing Compiler for Deep Learning
- Bubble-Up
- Applied Machine Learning at Facebook: A Datacenter Infrastructure Perspective
- GeePS
- Clipper: A Low-Latency Online Prediction Serving System
- Optimus
- Bubble-flux