DyPS: Dynamic Processor Switching for Energy-Aware Video Decoding on Multi-core SoCs
Yahia Benmoussa (Lab-STICC), Jalil Boukhobza (Lab-STICC), Eric Senn, (Lab-STICC), Djamel Benazzouz (MSS), Yassine Hadjadj-Aoul (INRIA - IRISA)

TL;DR
DyPS is a novel approach that dynamically switches between GPP and DSP processors for energy-efficient video decoding on multi-core SoCs, significantly reducing energy consumption while maintaining performance.
Contribution
This paper introduces DyPS, a new processor switching method for energy-aware video decoding, implemented on embedded Linux with GStreamer, improving energy efficiency in mobile devices.
Findings
Achieves 30% energy savings in video decoding
Maintains decoding performance despite energy savings
Validates effectiveness through a case study
Abstract
In addition to General Purpose Processors (GPP), Multicore SoCs equipping modern mobile devices contain specialized Digital Signal Processor designed with the aim to provide better performance and low energy consumption properties. However, the experimental measurements we have achieved revealed that system overhead, in case of DSP video decoding, causes drastic performances drop and energy efficiency as compared to the GPP decoding. This paper describes DyPS, a new approach for energy-aware processor switching (GPP or DSP) according to the video quality . We show the pertinence of our solution in the context of adaptive video decoding and describe an implementation on an embedded Linux operating system with the help of the GStreamer framework. A simple case study showed that DyPS achieves 30% energy saving while sustaining the decoding performance
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsVideo Coding and Compression Technologies · Low-power high-performance VLSI design · CCD and CMOS Imaging Sensors
