Hybrid dynamic thermal management based on statistical characteristics of multimedia applications

Abstract

Recently multimedia applications become one of the most popular applications in mobile devices such as wireless phones, PDAs, and laptops. However, typical mobile systems are not equipped with cooling components, which eventually causes critical thermal deficiencies. Although many low-power and low-temperature multimedia playback techniques have been proposed, they failed to provide QoS (Quality of Service) while controlling temperature due to the lack of proper understanding of multimedia applications. We propose Hybrid Dynamic Thermal Management (HDTM) which exploits thermal characteristics of both multimedia applications and systems. Specifically, we model application characteristics as the probability distribution of the number of cycles required to decode a frame. We also improve existing system thermal models by considering the effect of workload. This scheme finds an optimal clock frequency in order to prevent overheating with minimal performance degradation at runtime.The proposed scheme is implemented on Linux in a Pentium-M processor which provides variable clock frequencies. In order to evaluate the performance of the proposed scheme, we exploit three major codecs, namely MPEG-4, H.264/AVC and H.264/AVC streaming. Our results show that HDTM lowers the overall temperature by 15• C and the peak temperature by 20• C , while maintaining frame drop ratio under 0.2% compared to previous thermal management schemes such as feedback control DTM [8], Frame-based DTM [5] and GOP-based DTM [15].

Joint Work and Voltage/Frequency Scaling for Quality-Optimized Dynamic Thermal Management

“…Application-specific DTM methods have been extensively studied in the context of video encoding/decoding applications [3], [16]- [21]. In these studies, common DTM techniques were adopted and the timing characteristics of video encoding/decoding applications were used to guarantee real-time execution.…”

“…Another contribution of this paper is a more generic formulation and treatment of DWS for real-time applications. Previous DTM methods [3], [16]- [21] that have been applied to just real-time video encoding/decoding are dependent on application specifics. Therefore, they cannot easily be adapted to other real-time applications.…”

Heuristic-based real-time remote rendering of 3-D geometric model

“…Since most 3-D scenes not just consist of multiple objects, so, we extend our workload predicting model to estimate the workload of complex scenes with multiple objects. As we known, if a given mesh is rendered faster, more CPU time slices are consumed per unit time and more memory space is expended, but when the out frame rate reach a certain value it will can not show improvement in visual reality [7]. Thus, the goal of the CPU scheduler is to allot just enough CPU cycles to finish rendering the scene with least energy consumption.…”

Energy and thermal-aware video coding via encoder/decoder workload balancing

“…In [5], the authors develop a power-rate-distortion model for an MPEG encoder with complexity control parameters and solve an optimization problem to maximize video quality given power and framerate constraints. [6,7] both proposed dynamic thermal management schemes for MPEG decoding by allowing a certain amount of image/video quality loss. Each work made use of dynamic voltage/frequency scaling and some graceful performance degradation policies to ensure a thermally safe state of operation.…”