FaST-GShare: Enabling Efficient Spatio-Temporal GPU Sharing in Serverless Computing for Deep Learning Inference

TL;DR

FaST-GShare introduces a GPU sharing architecture for serverless deep learning inference that improves resource utilization, reduces costs, and guarantees service levels through spatio-temporal multiplexing and intelligent scheduling.

Contribution

It presents a novel FaST-GShare architecture with a dedicated manager, profiler, and scheduler for efficient, SLO-aware GPU sharing in serverless DL inference.

Findings

3.15x throughput improvement over time sharing

1.34x GPU utilization increase

3.13x SM occupancy enhancement

Abstract

Serverless computing (FaaS) has been extensively utilized for deep learning (DL) inference due to the ease of deployment and pay-per-use benefits. However, existing FaaS platforms utilize GPUs in a coarse manner for DL inferences, without taking into account spatio-temporal resource multiplexing and isolation, which results in severe GPU under-utilization, high usage expenses, and SLO (Service Level Objectives) violation. There is an imperative need to enable an efficient and SLO-aware GPU-sharing mechanism in serverless computing to facilitate cost-effective DL inferences. In this paper, we propose \textbf{FaST-GShare}, an efficient \textit{\textbf{Fa}aS-oriented \textbf{S}patio-\textbf{T}emporal \textbf{G}PU \textbf{Sharing}} architecture for deep learning inferences. In the architecture, we introduce the FaST-Manager to limit and isolate spatio-temporal resources for GPU multiplexing. In order to realize function performance, the automatic and flexible FaST-Profiler is proposed to profile function throughput under various resource allocations. Based on the profiling data and the isolation mechanism, we introduce the FaST-Scheduler with heuristic auto-scaling and efficient resource allocation to guarantee function SLOs. Meanwhile, FaST-Scheduler schedules function with efficient GPU node selection to maximize GPU usage. Furthermore, model sharing is exploited to mitigate memory contention. Our prototype implementation on the OpenFaaS platform and experiments on MLPerf-based benchmark prove that FaST-GShare can ensure resource isolation and function SLOs. Compared to the time sharing mechanism, FaST-GShare can improve throughput by 3.15x, GPU utilization by 1.34x, and SM (Streaming Multiprocessor) occupancy by 3.13x on average.