Evaluation of the Performance/Energy Overhead in DSP Video Decoding and its Implications

TL;DR

This paper evaluates the performance and energy overhead of DSP video decoding in mobile devices, revealing that communication overhead can significantly impact energy efficiency and sometimes favor general-purpose processors.

Contribution

It provides a detailed analysis of inter-processor communication overhead in DSP decoding and compares its impact to GPP decoding, highlighting scenarios where GPP may be more efficient.

Findings

DSP decoding overhead can reach 30% of total energy.

GPP decoding can outperform DSP in certain cases due to overhead.

Communication overhead significantly affects energy consumption.

Abstract

Video decoding is considered as one of the most compute and energy intensive application in energy constrained mobile devices. Some specific processing units, such as DSPs, are added to those devices in order to optimize the performance and the energy consumption. However, in DSP video decoding, the inter-processor communication overhead may have a considerable impact on the performance and the energy consumption. In this paper, we propose to evaluate this overhead and analyse its impact on the performance and the energy consumption as compared to the GPP decoding. Our work revealed that the GPP can be the best choice in many cases due to the a significant overhead in DSP decoding which may represents 30% of the total decoding energy.