PEW: Prediction-Based Early Dark Cores Wake-up Using Online Ridge Regression for Many-Core Systems

Abstract

Future many-core systems need to address the dark silicon problem, where some cores would be turned off to control the chip's thermal and power density, which effectively limits the performance gain from having a large number of processing cores. Task migration technique has been previously proposed to improve many-core system performance by moving tasks between active and dark cores. As task migration imposes system performance overhead due to the large wake-up latency of the dark cores, this paper proposes a prediction-based early wake-up (PEW) to reduce the dark cores' wake-up latency during task migration. A window-based online ridge regression (RR) is used as the prediction model. The prediction model uses the past window's thermal, power, and core status (i.e., active or dark) to predict the future core temperatures at run-time. If task migration is predicted in the next control period, the proposed PEW puts the dark cores in a power state with low wake-up latency. Thus, the proposed PEW reduces the time for the dark cores to start executing the tasks. The comparison results show that our proposed PEW reduces the completion time by up to 7.9% and 4.1% compared to non-early wake-up (NoEW) and a fixed threshold wake-up (FEW), respectively. It also shows that the proposed PEW increases the MIPS/Watt by up to 5.5% and 2.3% over NoEW and FEW, respectively. These results show that the proposed PEW improves the many-core system's overall performance in terms of reducing dark cores' wake-up latency and increasing the number of executed instructions per Watt.INDEX TERMS Dark silicon, many-core systems, task migration, dynamic voltage frequency scaling (DVFS), ridge regression, early wake-up, dark core wake-up latency.

CR-YOLOv8: Multiscale Object Detection in Traffic Sign Images

“…Automatic identification and detection of traffic signs on the road through image processing algorithms is an indispensable key link in driverless and ITS systems, providing an important basis for subsequent actions.In the realm of traffic sign detection, there is a considerable variation in the size, color, and shape of signs, significantly augmenting the complexity of the detection task. Consequently, the exploration of algorithms resilient to the diverse nature of traffic signs stands as a pivotal and enduring challenge in this domain [23]. In order to achieve the functions of indication, warning, restriction, and guidance, traffic signage is usually distinguished from its surroundings by eye-catching colors to enhance the identifiability of the signs.…”

Crowd Management Intelligence Framework: Umrah Use Case

“…Recent studies have used DL models to analyze the dynamic nature of crowd behavior such as visual motion segmentation [39] and motion classification [40]. The study in [39] proposed a stacked autoencoder network to segment complex crowd movement patterns during Hajj, and the researcher in [40] used a fully connected deep neural network to classify movement patterns in crowded scenes. In contrast to previous research, Alhazmi et al [37] introduced a webbased application for crowd management based on non-visual data collected using crowdsourcing techniques.…”

3D-DNaPE: Dynamic Neighbor-Aware Performance Enhancement for Thermally Constrained 3D Many-Core Systems

“…Mohammed et al [17], [18] introduced a dynamic thermal-aware performance optimization technique that uses task migration and DVFS to enhance the performance of thermally constrained manycore systems. Moreover, in [30], the researchers proposed a prediction-based early wake-up of dark cores to reduce the dark cores' wake-up latency and improve the overall performance of thermally constrained many-core systems. Several other researchers have also proposed techniques emphasizing dynamic power budgeting [31]- [33].…”