Experimental Evaluation and Modeling of Thermal Phenomena on Mobile Devices

Ferroni, Matteo; Nacci, Alessandro Antonio; Turri, Matteo; Santambrogio, Marco D.; Sciuto, D.

doi:10.1109/dsd.2015.20

Cited by 3 publications

(5 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the second case, the case in which only the smartphone's CPU cores are equipped with thermal sensors, we cannot directly observe the temperatures of the display chipsets. In this case, their temperatures can be approximately observed thanks to the following claim made by Ferroni et al [14]: Claim 2. For two hardware components, their temperature difference has a linear relationship with the physical distance between them.…”

Section: Online Thermal Predictionmentioning

confidence: 87%

“…Egilmez et al [13] presented a simple run-time method for estimating skin and screen temperature only with on-device thermal sensors and performance indicators. To describe the thermal interaction between the hardware components of the mobile devices, Ferroni et al [14] implemented a thermal prediction tool using a linear relationship between temperature difference and physical distance among hardware components. Paterna et al [15] proposed an exponential thermal prediction method considering the ambient condition and thermal coupling between mobile AP and the WiFi chipset.…”

Section: Thermal Prediction Methods For Mobile Devicesmentioning

confidence: 99%

“…In order to use Equation (21) for indirectly observing temperatures of a display chipset, in the offline phase, we find the product term d ij • K ij instead of separately finding d ij and K ij between a CPU and a display chipset. For this, our offline collected temperature data in Equation 15are used to calculate the average differences of average CPU cores' temperature ∑ C j=1 T c j (t)/C and the display chipsets' temperatures, i.e, Miracast chipset's temperature T Miracast (t) and HDMI chipset's temperature T HDMI (t) based on Ferroni et al's method [14]. Such calculated temperature differences are 22.51 • C and 14.95 • C, respectively and these values are used as the product terms d ij • K ij for online indirect observation of the temperatures of the Miracast chipset and HDMI chipset from the CPU temperature.…”

Section: Online Thermal Predictionmentioning

confidence: 99%

“…To protect microprocessors from thermal violations, there have been many studies on thermal prediction-based DTM (Dynamic Thermal Management) [6][7][8][9][10][11][12][13][14][15][16] to predict the temperature and then to lower the resource usage if the predicted temperature is at risk of exceeding the thermal threshold. Since those studies consider rapid temperature increases as abnormal and rare cases, they use simple solutions despite the risk of thermal violations.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Optimal Planning of Dynamic Thermal Management for NANS (N-App N-Screen) Services

et al. 2018

View full text Add to dashboard Cite

Existing multi-screening technologies have been limited to mirroring the current screen of the smartphone onto all the connected external display devices. In contrast, NANS (N-App N-Screen) technology is able to display different applications (N-App) on different multiple display devices (N-Screen) using only a smartphone. For such NANS services, this paper empirically shows that the thermal violation constraint is more critical than the battery life constraint. For preventing the thermal violation, the existing DTM (Dynamic Thermal Management) techniques cannot be used since they consider thermal violations as abnormal, and hence prevent them by severely throttling CPU frequencies resulting in serious QoS degradation. In NANS service scenarios it is normal to operate in high temperature ranges to continue services with acceptable QoS. Targeting such scenarios, we first propose a novel thermal prediction method specially designed for NANS services. Based on the novel thermal prediction method, we then propose a novel DTM technique called, “thermal planning” to provide sustainable NANS services with sufficiently high QoS without thermal violations.

show abstract

Section: Online Thermal Predictionmentioning

confidence: 87%

Section: Thermal Prediction Methods For Mobile Devicesmentioning

confidence: 99%

Section: Online Thermal Predictionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Optimal Planning of Dynamic Thermal Management for NANS (N-App N-Screen) Services

et al. 2018

View full text Add to dashboard Cite

show abstract

“…During the operating phase, certain models are built to optimize thermal behavior [12,40]. While, other models are used to study the reliability of the system by investigating the effects of internal and external temperatures on the chip [41], or on the lifetime of the system [42,43] to improve it [44].…”

Section: Temperature Modeling In Embeded Socsmentioning

confidence: 99%

Incremental Modeling and Monitoring of Embedded CPU-GPU Chips

Djedidi

Djeziri

2020

Processes

View full text Add to dashboard Cite

This paper presents a monitoring framework to detect drifts and faults in the behavior of the central processing unit (CPU)-graphics processing unit (GPU) chips powering them. To construct the framework, an incremental model and a fault detection and isolation (FDI) algorithm are hereby proposed. The reference model is composed of a set of interconnected exchangeable subsystems that allows it to be adapted to changes in the structure of the system or operating modes, by replacing or extending its components. It estimates a set of variables characterizing the operating state of the chip from only two global inputs. Then, through analytical redundancy, the estimated variables are compared to the output of the system in the FDI module, which generates alarms in the presence of faults or drifts in the system. Furthermore, the interconnected nature of the model allows for the direct localization and isolation of any detected abnormalities. The implementation of the proposed framework requires no additional instrumentation as the used variables are measured by the system. Finally, we use multiple experimental setups for the validation of our approach and also proving that it can be applied to most of the existing embedded systems.

show abstract