Abstract
Graphics processing units (GPUs) are extensively used as accelerators across multiple application domains, ranging from general purpose applications to neural networks, and cryptocurrency mining. The initial utilization paradigm for GPUs was one application accessing all the resources of the GPU. In recent years, time sharing is broadly used among applications of a GPU, nevertheless, spatial sharing is not fully explored. When concurrent applications share the computational resources of a GPU, performance can be improved by eliminating idle resources. Additionally, the incorporation of GPUs in embedded and mobile devices increases the demand for power efficient computation due to battery limitations. In this article, we present an allocation methodology for streaming multiprocessors (SMs). The presented methodology works for two concurrent applications on a GPU and determines an allocation scheme that will provide power efficient application execution, combined with improved GPU performance. Experimental results show that the developed methodology yields higher throughput while achieving improved power efficiency, compared to other SM power-aware and performance-aware policies. If the presented methodology is adopted, it will lead to higher performance of applications that are concurrently executing on a GPU. This will lead to a faster and more efficient acceleration of execution, even for devices with restrained energy sources.
Highlights
Application demands for computational resources continuously rise
To obtain power measurements for the applications of the experiment we used GPUWattch [24], which is a simulator that integrates with general purpose GPUs (GPGPUs)-Sim and provides power consumption profiles for the simulations run on GPGPU-Sim
We provide details about the Graphics processing units (GPUs) configuration we used during the experiments in Tables 1 and 2
Summary
Application demands for computational resources continuously rise. To meet these demands, software engineers used to take advantage of the improvement on hardware technology, e.g., smaller transistor dimensions, higher clock frequencies, and chips with numerous processing cores.high transistor density and shrinking transistor dimensions have reached a point where improvement in performance cannot be further achieved. Application demands for computational resources continuously rise. To meet these demands, software engineers used to take advantage of the improvement on hardware technology, e.g., smaller transistor dimensions, higher clock frequencies, and chips with numerous processing cores. To increase performance for parallel applications, system engineers have turned to the usage of hardware accelerators. Graphics processing units (GPUs) have been extensively utilized as accelerators, providing significant improvements in performance. GPUs were initially developed to accelerate graphics rendering. Since their first introduction, the necessary tools and frameworks have been developed, allowing programmers to leverage the computational power of GPUs in various application domains. Some examples of areas where GPUs are being used are in the acceleration of neural networks, autonomous cars, and cryptocurrency mining
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
