Contention-Aware GPU Partitioning and Task-to-Partition Allocation for Real-Time Workloads

TL;DR

This paper introduces contention-aware GPU partitioning and task allocation techniques for real-time workloads, improving scheduling efficiency by considering kernel interference in multi-cluster GPU architectures.

Contribution

It proposes novel GPU partitioning and task-to-partition allocation methods that account for kernel interference, enhancing real-time scheduling on modern GPU architectures.

Findings

Proposed techniques outperform classical non-partitioned approaches

Effective interference quantification improves scheduling accuracy

Partitioning increases GPU utilization for real-time tasks

Abstract

In order to satisfy timing constraints, modern real-time applications require massively parallel accelerators such as General Purpose Graphic Processing Units (GPGPUs). Generation after generation, the number of computing clusters made available in novel GPU architectures is steadily increasing, hence, investigating suitable scheduling approaches is now mandatory. Such scheduling approaches are related to mapping different and concurrent compute kernels within the GPU computing clusters, hence grouping GPU computing clusters into schedulable partitions. In this paper we propose novel techniques to define GPU partitions; this allows us to define suitable task-to-partition allocation mechanisms in which tasks are GPU compute kernels featuring different timing requirements. Such mechanisms will take into account the interference that GPU kernels experience when running in overlapping time windows. Hence, an effective and simple way to quantify the magnitude of such interference is also presented. We demonstrate the efficiency of the proposed approaches against the classical techniques that considered the GPU as a single, non-partitionable resource.