Supporting Preemptive Task Executions and Memory Copies in GPGPUs

Başaran, Can; Kang, Kyoung-Don

doi:10.1109/ecrts.2012.15

Cited by 78 publications

(39 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ways in which GPUs are managed and scheduled differ greatly from CPUs. This has spurred research on supporting GPUs in real-time systems [1,2,3,4,5,6,7,8,9,10]. Still, few have explored multiprocessor, multi-GPU real-time systems.…”

Section: Introductionmentioning

confidence: 99%

Exploring the Multitude of Real-Time Multi-GPU Configurations

Elliott

Anderson

2014

2014 IEEE Real-Time Systems Symposium

View full text Add to dashboard Cite

Abstract-Motivated by computational capacity and power efficiency, techniques for integrating graphics processing units (GPUs) into real-time systems have become an active area of research. While much of this work has focused on single-GPU systems, multiple GPUs may be used for further benefits. Similar to CPUs in multiprocessor systems, GPUs in multi-GPU systems may be managed using partitioned, clustered, or global methods, independent of CPU organization. This gives rise to many combinations of CPU/GPU organizational methods that, when combined with additional GPU management options, results in thousands of "reasonable" configuration choices. In this paper, we explore real-time schedulability of several categories of configurations for multiprocessor, multi-GPU systems that are possible under GPUSync, a recently proposed highly configurable real-time GPU management framework. Our analysis includes the careful consideration of GPU-related overheads. We show system configuration strongly affects realtime schedulability. We also identify which configurations offer the best schedulability in order to guide the implementation of GPU-based real-time systems and future research.

show abstract

Section: Introductionmentioning

confidence: 99%

Exploring the Multitude of Real-Time Multi-GPU Configurations

Elliott

Anderson

2014

2014 IEEE Real-Time Systems Symposium

View full text Add to dashboard Cite

show abstract

“…In its absence, tricks, like cutting longer requests into smaller pieces, have been shown to enhance GPU interactivity, at least for mutually cooperative applications [6]. True hardware preemption support would save state and safely context-switch from an ongoing request to the next in the GPU queue.…”

Section: Hardware Preemption Supportmentioning

confidence: 99%

Disengaged scheduling for fair, protected access to fast computational accelerators

Menychtas

Shen

Scott

2014

Proceedings of the 19th International Conference on Architectural Support for Programming Languages and Operating Systems

View full text Add to dashboard Cite

Today's operating systems treat GPUs and other computational accelerators as if they were simple devices, with bounded and predictable response times. With accelerators assuming an increasing share of the workload on modern machines, this strategy is already problematic, and likely to become untenable soon. If the operating system is to enforce fair sharing of the machine, it must assume responsibility for accelerator scheduling and resource management.Fair, safe scheduling is a particular challenge on fast accelerators, which allow applications to avoid kernel-crossing overhead by interacting directly with the device. We propose a disengaged scheduling strategy in which the kernel intercedes between applications and the accelerator on an infrequent basis, to monitor their use of accelerator cycles and to determine which applications should be granted access over the next time interval.Our strategy assumes a well defined, narrow interface exported by the accelerator. We build upon such an interface, systematically inferred for the latest Nvidia GPUs. We construct several example schedulers, including Disengaged Timeslice with overuse control that guarantees fairness and Disengaged Fair Queueing that is effective in limiting resource idleness, but probabilistic. Both schedulers ensure fair sharing of the GPU, even among uncooperative or adversarial applications; Disengaged Fair Queueing incurs a 4% overhead on average (max 18%) compared to direct device access across our evaluation scenarios.

show abstract

“…This situation can cause unfairness between multiple kernels and significantly deteriorate the system responsiveness. Existing GPU scheduling methods address this issue by either killing a long running kernel [Menychtas et al 2014] or providing a kernel split tool [Basaran and Kang 2012;Zhou et al 2015;Margiolas and O'Boyle 2016]. The Pascal architecture allows GPU kernels to be interrupted at instruction-level granularity by saving and restoring each GPU context to and from the GPU's DRAM.…”

Section: Algorithms For Scheduling a Single Gpumentioning

confidence: 99%

GPU Virtualization and Scheduling Methods

2017

View full text Add to dashboard Cite

The integration of graphics processing units (GPUs) on high-end compute nodes has established a new accelerator-based heterogeneous computing model, which now permeates high performance computing. The same paradigm nevertheless has limited adoption in cloud computing or other large-scale distributed computing paradigms. Heterogeneous computing with GPUs can benefit the Cloud by reducing operational costs and improving resource and energy efficiency. However, such a paradigm shift would require effective methods for virtualizing GPUs, as well as other accelerators. In this survey paper, we present an extensive and in-depth survey of GPU virtualization techniques and their scheduling methods. We review a wide range of virtualization techniques implemented at the GPU library, driver, and hardware levels. Furthermore, we review GPU scheduling methods that address performance and fairness issues between multiple virtual machines sharing GPUs. We believe that our survey delivers a perspective on the challenges and opportunities for virtualization of heterogeneous computing environments.

show abstract

Supporting Preemptive Task Executions and Memory Copies in GPGPUs

Cited by 78 publications

References 17 publications

Exploring the Multitude of Real-Time Multi-GPU Configurations

Exploring the Multitude of Real-Time Multi-GPU Configurations

Disengaged scheduling for fair, protected access to fast computational accelerators

GPU Virtualization and Scheduling Methods

Contact Info

Product

Resources

About