Real-Time Scheduling Using GPUs – Advanced and More Accurate Proof of Feasibility

Peter Fodrek, Ludovit Farkas, Michal Blaho, Martin Foltin, Juraj Hnat, Tomas Murgas
Institute of Control and Industrial Informatics, Faculty of Electrical Engineering and Information Technology, Slovak University of Technogy, Bratislava 812 19,Slovak republic
Journal of Communication and Computer 9, 863-871, 2012


   title={Real-Time Scheduling Using GPUs-Advanced and More Accurate Proof of Feasibility},

   author={Fodrek, P. and Farkas, L’. and Blaho, M. and Foltin, M. and Hn{‘a}t, J. and Murgas, T.},

   journal={Journal of Communication and Computer},





Download Download (PDF)   View View   Source Source   



This paper will report our evaluation to use OpenCL as a platform for hard real-time scheduling. Especially, we have evaluated which types of tasks are faster on GPGPU than on CPU. We have investigated computational tasks, memory intensive tasks (especially tasks using low latency GDDR memory) and disk intensive tasks. This study is the part of a larger research program to design an innovative Linux scheduler subsystem that runs on GPGPU and schedules tasks running on GPGPU as well as on CPU. Based on the results we obtained from benchmarking various types of tasks we found out that some of them are faster on GPGPU than on CPU and therefore should preferably be executed on GPGPU. Preliminary data suggest that we can expect a speed up of up to 10-fold with respect to execution time and latency. Main problems of this approach is videoRAM to RAM data transfer. This will be solved via hardware vendor Advanced Micro Devices in 2014 as their road-map form February 2, 2012 shows. Additionally we present performance of GPU against multi-thread and single-threaded CPU case.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2021 hgpu.org

All rights belong to the respective authors

Contact us: