Due to limited resources on edge and different characteristics of deep neural network (DNN) models, it is a big challenge to optimize DNN inference performance in terms of energy consumption and end-to-end latency. In addition to dynamic voltage frequency scaling (DVFS) technique, edge-cloud architecture provides a collaborative approach for efficient DNN inference. However, current edge-cloud collaborative inference methods have not optimized various compute resources on edge devices. Thus, we propose DVFO, a novel DVFS-enabled edge-cloud collaborative inference framework, which co-optimizes DVFS and offloading parameters via deep reinforcement learning (DRL). Specifically, DVFO automatically co-optimizes 1) the CPU, GPU and memory frequencies of edge devices, and 2) the offloaded feature map. In addition, it leverages a thinking-while-moving concurrent mechanism to accelerate the DRL learning process, and a spatial-channel attention mechanism to identify the less important DNN feature map for efficient offloading. This approach improves inference performance for different DNN models under various edge-cloud network conditions. Extensive evaluations using two datasets and six widely-deployed DNN models on five heterogeneous edge devices show that DVFO significantly reduces the energy consumption by 33% on average, compared to state-of-the-art schemes. Moreover, DVFO achieves up to 28.6%∼59.1% end-to-end latency reduction, while maintaining accuracy within 1% loss on average.

Nowadays, mobile edge computing (MEC) appears as a new computing paradigm with its ability to utilize the computing power of both local devices and edge servers. In MEC, edge pricing and resource allocation (EPRA) are two important problems. Edge servers make a profit by selling computing services to users. To maximize their revenue, they need to determine an appropriate price for each user, and decide the amount of resources allocated to each user. However, none of the existing works consider the effect of users’ task assignment strategy on the revenue of the edge. In fact, EPRA will affect the users’ task offloading decision, as they expect to minimize their total cost. In turn, the users’ decision will also influence the revenue of the edge. Therefore, the interaction between mobile users and edge servers should be considered carefully and the interests of both sides need to be maximized simultaneously. In this article, we model the interaction between the two sides as a Stackelberg game. First, given a specified EPRA strategy, we derive a near-optimal task assignment strategy for each user to minimize the total cost based on a greedy algorithm UTA-G. Then, by applying the backward induction method, two pricing and resource allocation schemes with different granularity, i.e., user-granular EPRA and task-granular EPRA are proposed to bring higher revenue to the edge. Experimental results demonstrate that all the proposed algorithms can have good performance in task-intensive, resource-deficient, and workload-heavy scenarios.

Crop phenology is an important attribute of crops, not only reflecting the growth and development of crops, but also affecting crop yield. By observing the phenological stages, agricultural production losses can be reduced and corresponding systems and plans can be formulated according to their changes, having guiding significance for agricultural production activities. Traditionally, crop phenological stages are determined mainly by manual analysis of remote sensing data collected by UAVs, which is time-consuming, labor-intensive, and may lead to data loss. To cope with this problem, this paper proposes a deep-learning-based method for rice phenological stage recognition. Firstly, we use a weather station equipped with RGB cameras to collect image data of the whole life cycle of rice and build a dataset. Secondly, we use object detection technology to clean the dataset and divide it into six subsets. Finally, we use ResNet-50 as the backbone network to extract spatial feature information from image data and achieve accurate recognition of six rice phenological stages, including seedling, tillering, booting jointing, heading flowering, grain filling, and maturity. Compared with the existing solutions, our method guarantees long-term, continuous, and accurate phenology monitoring. The experimental results show that our method can achieve an accuracy of around 87.33%, providing a new research direction for crop phenological stage recognition.

Can we reduce the search cost of Neural Architecture Search (NAS) from days down to only a few hours? NAS methods automate the design of Convolutional Networks (ConvNets) under hardware constraints and they have emerged as as a key component of cloud AutoML services. However, the NAS problem remains challenging due to the combinatorially large design space and the significant search time (at least 200 GPU-hours). In this work, we alleviate the NAS search cost down to less than 3 hours, while achieving state-of-the-art image classification results under mobile latency constraints. We propose a novel differentiable NAS formulation, namely Single-Path NAS, that uses one single-path over-parameterized ConvNet to encode all architectural decisions based on shared convolutional kernel parameters, hence drastically decreasing the search overhead. Single-Path NAS achieves state-of-the-art top-1 ImageNet accuracy (75.62%), hence outperforming existing mobile NAS methods in similar latency settings (∼80ms). In particular, we enhance the accuracy-runtime trade-off in differentiable NAS by treating the Squeeze-and-Excitation path as a fully searchable operation with our novel single-path encoding. Our method has an overall cost of only 8 epochs (24 TPU-hours), which is up to 5,000× faster compared to prior work. Moreover, we study how different NAS formulation choices affect the performance of the designed ConvNets. Furthermore, we exploit the efficiency of our method to answer an interesting question: instead of empirically tuning the hyperparameters of the NAS solver (as in prior work), can we automatically find the hyperparameter values that yield the desired accuracy-runtime trade-off (e.g., target runtime for different platforms)? We view our extensive experimental results as a valuable exploration for NAS-based cloud AutoML services, and we open-source our entire codebase at: https://github.com/dstamoulis/single-path-nas.

The satellite-terrestrial integrated networks based on edge computing can provide computation offloading service to terminal devices in remote areas. However, it faces various limitations, including satellite energy consumption, computation delay, and environmental dynamics, etc. In this paper, we propose a satellite-terrestrial integrated cloud and edge computing network (STCECN) architecture, including satellite layer, terrestrial layer and cloud center, where computing resources exist in multi-layer heterogeneous edge computing clusters. The optimization of system delay and energy consumption is defined as a mixed integer programming problem. Moreover, we present a deep reinforcement learning based computation offloading decision algorithm that can adapt to the dynamics and variability of satellite networks. A dual-network computing offloading decision method for delay and energy consumption based on deep reinforcement learning offloading (DRLO) is proposed, including deep convolutional network update method, quantization strategy, and bandwidth resource allocation. Meanwhile, integrating decision-making bias adjustment strategies based on past experience to address the problem of pseudo packet loss caused by satellite network switching. The simulation results indicate that the proposed method performs almost consistently with traditional heuristic algorithms, with only 20% of the time consumption of the latter, and the number of pseudo packet loss also decreases to the original 10-20%.

The use of big data leads to higher demands for hyperscale data centers (HDCs) in terms of the scale and quantity required for data storage and processing. Before the construction of an HDC, it is necessary to comprehensively analyze the economic budget according to the energy requirements and potential energy cost. We propose a global energy consumption prediction framework based on the power usage effectiveness (PUE) calculation that considers all heat sources and power consumption. The framework integrates physical models and a statistical framework that combines IT equipment energy consumption and data center energy consuming predictions. Furthermore, the framework provides a method to calculate the carbon emissions and electricity cost of the data center. Using hourly meteorological data as climate parameters, combined with a limited range of energy parameters, the annual PUE values of 60 regions were estimated, and a further analysis of the Carbon Usage Effectiveness (CUE) and electricity costs in China was conducted as an example. Based on experimental validation and an evaluation of real-time data, our framework can predict the overall energy consumption of HDCs effectively, filling a gap in HDC research in the Asia-Pacific region and providing a basis for HDC feasibility analysis.