Given recent algorithm, software, and hardware innovation, computing has enabled a plethora of new applications. As becomes increasingly ubiquitous, however, so does its environmental impact. This paper brings the issue to attention computer-systems researchers. Our analysis, built on industry-reported characterization, quantifies effects in terms carbon emissions. Broadly, emissions have two sources: operational energy consumption, manufacturing infrastructure. Although from former are...

Given recent algorithm, software, and hardware innovation, computing has enabled a plethora of new applications. As becomes increasingly ubiquitous, however, so does its environmental impact. This article brings the issue to attention computer-systems researchers. Our analysis, built on industry-reported characterization, quantifies effects in terms carbon emissions. Broadly, emissions have two sources: operational energy consumption, manufacturing infrastructure. Although from former are...

Deep learning inference is increasingly run at the edge. As programming and system stack support becomes mature, it enables acceleration opportunities in a mobile system, where performance envelope scaled up with plethora of programmable co-processors. Thus, intelligent services designed for users can choose between running on CPU or any co-processors exploiting connected systems such as cloud nearby, locally system. By doing so, these scale out increase energy efficiency edge systems. This...

Federated learning enables a cluster of decentralized mobile devices at the edge to collaboratively train shared machine model, while keeping all raw training samples on device. This approach is demonstrated as practical solution mitigate risk privacy leakage. However, enabling efficient FL deployment challenging because non-IID data distribution, wide system heterogeneity and stochastic-varying runtime effects in field. paper jointly optimizes time-to-convergence energy efficiency...

Recent mobile devices adopt high-performance processors to support various functions. As a side effect, higher performance inevitably leads power density increase, eventually resulting in thermal problems. In order alleviate the problems, off-the-shelf rely on dynamic voltage-frequency scaling (DVFS)-based management (DTM) schemes. Unfortunately, DVFS-based DTM schemes, an excessive number of operations worsen not only but also efficiency. this paper, we propose temperature-aware DVFS scheme...

Off-the-shelf embedded systems have adopted heterogeneous multi-core processors which high-performance big cores and low-power small cores. Though there are two different types of in processors, conventional DVFS (Dynamic Voltage Frequency Scaling)-based DTM Thermal Management) techniques do not utilize the to cool down hot Rather, they primarily reduce voltage frequency cores, leading performance degradation. In this article, we propose a novel adaptive framework for utilizes prevent Our...

Thin films of gadolinium-doped ceria (GDC) nanoparticles were fabricated as electrolytes for low-temperature solid oxide fuel cells (SOFCs) by combining electrophoretic deposition (EPD) and the successive ionic layer adsorption reaction-air spray plus (SILAR-A+) method. The Ce1−xGdxO2−x/2 solution was synthesized using hydrothermal (HY) solid-state (SS) procedures to produce high-quality GDC suitable EPD fabrication. crystalline structure, cell parameters, phases products analyzed X-ray...

To improve mobile experience of users, recent devices have adopted powerful processing units (CPUs and GPUs). Unfortunately, the often consume a considerable amount energy, which in turn shortens battery life devices. For energy reduction units, adopt management techniques based on software, especially OS (Operating Systems), as well hardware. In this survey paper, we summarize OS-level for units. We categorize into three parts, according to main operations summarized techniques: 1)...

Recent smart devices have adopted heterogeneous multi-core processors which high-performance big cores and low-power small cores. Unfortunately, the conventional task scheduler for does not provide appropriate amount of CPU resources multimedia applications (whose QoS is important to users), resulting in energy waste; it often executes non-multimedia on same core. In this paper, we propose an advanced processors, provides applications. Our proposed isolates from at runtime, exploiting fact...

A case study reveals that energy savings do not always translate to longer smartphone battery life and evaluating any plan must be based on consumption, used.

As Wi-Fi handheld devices providing a variety of data and multimedia services become more widespread, as the global demand for internet access everywhere rapidly increases, energy-efficient network mobility (NEMO) technology is attracting great attention with respect to latest research in mobile networking. The previous relating networks networking has focused on routing protocol or topology control. In NEMO, cost-efficient design reduces unnecessary costs because signalling packet delivery,...

As GPU power consumption in smartphones increases with more advanced graphic performance, it becomes essential to estimate accurately. The conventional model assumes, simply, that a consumes constant when turned on; however, this is no longer true for recent smartphone GPUs. In paper, we propose an accurate smartphones, considering newly adopted dynamic voltage and frequency scaling. For the proposed model, our evaluation results show error rate system estimation as low 2.9%, on average,...

To prolong battery life of mobile devices, image processing applications often exploit offloading techniques which run some or all the computations on remote servers. Unfortunately, existing do not consider fact that data transmission time and energy consumption wireless network interfaces exponentially increase when signal strength decreases. In this paper, we propose an adaptive for applications, considers strength. improve performance efficiency offloading, also to adaptively local...

Endoscopic surgical technology has been developing rapidly. Although several successful endoscopic closure devices have already introduced, only a few of them demonstrated improvements in strength and reproducibility over conventional clip closures or hand sutures. The objective this study was to test the feasibility novel successive suturing device (SSD) by measuring reproducibility.Porcine stomach models were used study. Endoclips, full-thickness sutures, SSD sutures close perforation...

Recently, mobile devices have employed heterogeneous multi-core processors which consist of highperformance big cores and low-power small cores. In processors, the conventional DVFS (Dynamic Voltage Frequency Scaling)-based DTM Thermal Management) is still adopted; it does not actively utilize to resolve thermal problem. this paper, we propose M-DTM (Migration-based DTM) for processors. case emergency cores, migrates applications instead lowering voltage frequency way, allows more time run...

A new alkoxide precursor, Hf(mp)4 [mp = 3-methyl-3-pentoxide, OC(CH3)(C2H5)2] has been employed in the atomic layer deposition (ALD) of HfO2 thin films using water (H2O) as oxygen source. The self-limiting ALD process via alternate surface reactions and H2O is confirmed from thickness measurements grown with varied supply time number Hf(mp)4/H2O cycles. temperature window for this precursor found to be between 250 350 °C. Under optimal reaction conditions, growth rate ∼0.9 Å per cycle....

To prolong battery life of mobile devices, applications often exploit offloading techniques which run computations on remote servers. Unfortunately, the existing do not consider fact that data transmission time and energy consumption wireless network interfaces exponentially increase when signal strength decreases. In this paper, we propose an adaptive technique considers strength. Our estimates gain (reduced computation devices) loss (increased interfaces) depending Based estimated loss,...

Virtualized data centers usually consist of heterogeneous servers which have different specifications (performance). Though there are a number unused with performance in such centers, conventional DVFS (Dynamic Voltage and Frequency Scaling)-based DTM Thermal Management) techniques do not exploit the to cool down hot servers. In this paper, we propose novel technique adaptively exploits external computing resources (unused performance) as well internal CPU cores server) available centers....