A study of mobile device utilization

Cao, Gao; Gutierrez, Anthony; Rajan, M. Sundar; Dreslinski, Ronald G.; Mudge, Trevor; Wu, Carole-Jean

doi:10.1109/ispass.2015.7095808

Cited by 50 publications

(14 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Gao et al [8] observe that offloading applications with high thread/data-level parallelism onto GPUs or accelerators can offer better energy efficiency. The speed of user interaction (e.g.…”

Section: Related Work 21 Smartphone Workload Characterisationmentioning

confidence: 99%

Impact of memory frequency scaling on user-centric smartphone workloads

Mendis

Chen

Indrusiak

et al. 2018

Proceedings of the 33rd Annual ACM Symposium on Applied Computing

View full text Add to dashboard Cite

Improving battery life in mobile phones has become a top concern with the increase in memory and computing requirements of applications with tough quality-of-service needs. Many energy-efficient mobile solutions vary the CPU and GPU voltage/frequency to save power consumption. However, energy-aware control over the memory bus connecting the various on-chip subsystems has had much less interest. This measurement-based study first analyse the CPU, GPU and memory cost (i.e. product of utilisation and frequency) of user-centric smartphone workloads. The impact of memory frequency scaling on power consumption and qualityof-service is also measured. We also present a preliminary analysis into the frequency levels selected by the different default governors of the CPU/GPU/memory components. We show that an interdependency exists between the CPU and memory governors and that it may cause unnecessary increase in power consumption, due to interference with the CPU frequency governor. The observations made in this measurement-based study can also reveal some design insights to system designers.

show abstract

Section: Related Work 21 Smartphone Workload Characterisationmentioning

confidence: 99%

Impact of memory frequency scaling on user-centric smartphone workloads

Mendis

Chen

Indrusiak

et al. 2018

Proceedings of the 33rd Annual ACM Symposium on Applied Computing

View full text Add to dashboard Cite

show abstract

“…To explore the opportunity, several studies have examined how much parallelism is there in mobile applications [13][14][15].…”

Section: Workload Analysismentioning

confidence: 99%

“…Gao et al analyze how the multiple cores are utilized in mobile devices using TLP (Thread Level Parallelism) [13]. They report that TLP of the most mobile applications including browser, map, music, and games is less than 2 on average, meaning that the number of active cores used by the majority of applications is below 2.…”

Section: Workload Analysismentioning

confidence: 99%

See 1 more Smart Citation

An Adaptive and Integrated Low-Power Framework for Multicore Mobile Computing

Choi

Jung

Choi

et al. 2017

Mobile Information Systems

View full text Add to dashboard Cite

Employing multicore in mobile computing such as smartphone and IoT (Internet of Things) device is a double-edged sword. It provides ample computing capabilities required in recent intelligent mobile services including voice recognition, image processing, big data analysis, and deep learning. However, it requires a great deal of power consumption, which causes creating a thermal hot spot and putting pressure on the energy resource in a mobile device. In this paper, we propose a novel framework that integrates two well-known low-power techniques, DPM (Dynamic Power Management) and DVFS (Dynamic Voltage and Frequency Scaling) for energy efficiency in multicore mobile systems. The key feature of the proposed framework is adaptability. By monitoring the online resource usage such as CPU utilization and power consumption, the framework can orchestrate diverse DPM and DVFS policies according to workload characteristics. Real implementation based experiments using three mobile devices have shown that it can reduce the power consumption ranging from 22% to 79%, while affecting negligibly the performance of workloads.

show abstract

“…Evolution of multicore architectures helped reaching those requirements, however, also results in severe power consumption issu when many cores are active simultaneously [2]. Besides that, it may appear that using more cores at the same time does not necessarily lead to better performance or better user-experience because of many realistic constraints [3]. Hence, there is significant room to 2 save power for mobile devices while maintaining good performance level by carefully managing the available processing cores [14].…”

Section: Introductionmentioning

confidence: 99%

MobiCore: An adaptive hybrid approach for power-efficient CPU management on Android devices

Broyde

Nixon

Chen

et al. 2017

2017 30th IEEE International System-on-Chip Conference (SOCC)

View full text Add to dashboard Cite

Smartphones are becoming essential devices used for various types of applications in our daily life. To satisfy the ever-increasing performance requirement, the number of CPU cores in a phone keeps growing, which imposes a great impact on its power consumption. This work presents a series of analysis to understand how the current Android resource management policy adjusts CPU features. Our results indicate a significant improvement margin for CPU power efficiency in modern Android smartphones. We then propose MobiCore -a power-efficient CPU management scheme that can optimize the use of Dynamic and Frequency Voltage Scaling (DVFS) and the Dynamic Core Scaling (DCS) techniques with a sensitive control on CPU bandwidth. The measurements on the real systems prove that MobiCore can achieve substantial Thanks to Dr. El-Nokali who brought me here. For his support and availability for my endless questions. For his trust in me as well as for the great opportunity he offered me.x iMy advisor, Dr. Chen giving me the opportunity to do research on his behalf. As well as for the time he spent for me.Sandy Weisberg for her constant support for all international students. 1 INTRODUCTIONDue to battery constraints, energy efficiency is, today, the main concern in mobile devices, such as the smartphone. Given the decreasing size of hardware components, technology helps us to combine more and more components together in a smaller environment. The emerging market for multi-core processors is giving new opportunities in terms of resources management as we have more options in our hand such as variability which enables us to get the exact output from those hardware components. From single core architecture, we went through two-core, four-core, octa-core and now reaching deca-core implementation in modern mobile devices [17]. Such new systems and architectures raise new challenges and new opportunities for energy consumption and performance level.Multi-core processors are now common design in embedded systems and they add a whole new dimension to the given opportunities for management in mobile devices. It now becomes essential for mobile applications to fully exploit the potentials of this available embedded hardware resource [15]. Especially for the ever increasing high-demand in computing performance in mobile devices which is going higher each year [15]. Evolution of multicore architectures helped reaching those requirements, however, also results in severe power consumption issu when many cores are active simultaneously [2]. Besides that, it may appear that using more cores at the same time does not necessarily lead to better performance or better user-experience because of many realistic constraints [3]. Hence, there is significant room to 2 save power for mobile devices while maintaining good performance level by carefully managing the available processing cores [14]. BACKGROUNDIn modern multi-core processors embedded in mobile devices, the OS system (here Android OS)provides multiple policies to control energy consumption....

show abstract

A study of mobile device utilization

Cited by 50 publications

References 27 publications

Impact of memory frequency scaling on user-centric smartphone workloads

Impact of memory frequency scaling on user-centric smartphone workloads

An Adaptive and Integrated Low-Power Framework for Multicore Mobile Computing

MobiCore: An adaptive hybrid approach for power-efficient CPU management on Android devices

Contact Info

Product

Resources

About