Search

With the rapid development of Internet of Things (IoT) and next-generation communication technologies, resource-constrained mobile devices fail to meet the demand of resource-hungry and compute-intensive applications. To cope with this challenge, with the assistance of Mobile Edge Computing (MEC), offloading complex tasks from mobile devices to edge cloud servers or central cloud servers can reduce the computational burden of devices and improve the efficiency of task processing. However, it is difficult to obtain optimal offloading decisions by conventional heuristic optimization methods, because the decision-making problem is usually NP-hard. In addition, there are shortcomings in using intelligent decision-making methods, e.g., lack of training samples and poor ability of migration under different MEC environments. To this end, we propose a novel offloading algorithm named MR-DRO, consisting of a Meta-Reinforcement Learning (meta-RL) model, which improves the migration ability of the whole model, and a Deep Reinforcement Learning (DRL) model, which combines multiple parallel Deep Neural Networks (DNNs) to learn from historical task offloading scenarios. Simulation results demonstrate that our approach can effectively and efficiently generate near-optimal offloading decisions in IoT environments with edge and cloud collaboration, which further improves the computational performance and has strong portability when making offloading decisions.

Cloud computing has been regarded as a successful paradigm for IT industry by providing benefits for both service providers and customers. In spite of the advantages, cloud computing also suffers from distinct challenges, and one of them is the inefficient resource provisioning for dynamic workloads. Accurate workload predictions for cloud computing can support efficient resource provisioning and avoid resource wastage. However, due to the high-dimensional and high-variable features of cloud workloads, it is difficult to predict the workloads effectively and accurately. The current dominant work for cloud workload prediction is based on regression approaches or recurrent neural networks, which fail to capture the long-term variance of workloads. To address the challenges and overcome the limitations of existing works, we proposed an e fficient supervised learning-based D eep N eural Network ( esDNN ) approach for cloud workload prediction. First, we utilize a sliding window to convert the multivariate data into a supervised learning time series that allows deep learning for processing. Then, we apply a revised Gated Recurrent Unit (GRU) to achieve accurate prediction. To show the effectiveness of esDNN, we also conduct comprehensive experiments based on realistic traces derived from Alibaba and Google cloud data centers. The experimental results demonstrate that esDNN can accurately and efficiently predict cloud workloads. Compared with the state-of-the-art baselines, esDNN can reduce the mean square errors significantly, e.g., 15%. rather than the approach using GRU only. We also apply esDNN for machines auto-scaling, which illustrates that esDNN can reduce the number of active hosts efficiently, thus the costs of service providers can be optimized.

As a distributed computing paradigm, edge computing has become a key technology for providing timely services to mobile devices by connecting Internet of Things (IoT), cloud centers and other facilities. By offloading compute-intensive tasks from IoT devices to edge/cloud servers, the communication and computation pressure caused by the massive data in industrial IoT can be effectively reduced. In the process of computation offloading in edge computing, it is critical to dynamically make optimal offloading decisions to minimize the delay and energy consumption spent on the devices. Although there are a large number of task offloading-decision models, how to measure and evaluate the quality of different models and configurations is crucial. In this paper, we propose a novel simulation platform named ChainFL, which can build an edge computing environment among IoT devices while being compatible with Federated Learning and Blockchain technologies to better support the embedding of security-focused offloading algorithms. ChainFL is lightweight and compatible, and it can quickly build complex network environments by connecting devices of different architectures. Moreover, due to its distributed nature, ChainFL can also be deployed as a federated learning platform across multiple devices to enable federated learning with high security due to its embedded blockchain. Finally, we validate the versatility and effectiveness of ChainFL 2 by embedding a complex offloading decision model in the platform, as well as deploying it in an industrial IoT environment with security risks.

Big data frameworks such as Apache Spark are becoming prominent to perform large-scale data analytics jobs. However, local or on-premise computing resources are often not sufficient to run these jobs. Therefore, public cloud resources can be hired on a pay-per-use basis from the cloud service providers to deploy a Spark cluster entirely on the cloud. Nevertheless, using only cloud resources can be costly. Hence, now-a-days, both local and cloud resources are used together to deploy a hybrid cloud computing cluster. However, scheduling jobs in a cluster deployed on hybrid cloud is challenging in the presence of various Service-Level Agreement (SLA) demands such as cost minimization and job deadline guarantee. Most of the existing works either consider a public or a locally deployed cluster and mainly focus on improving job performance in the cluster. In this paper, we propose efficient scheduling algorithms that leverage different cost models in a hybrid cloud deployed cluster to optimize the Virtual Machine (VM) usage cost for both local and cloud resources and maximize the job deadline meet percentage. The results show that our proposed algorithms are highly scalable and reduce the cost of VM usage of a hybrid cluster for up to 20%.

The advent of vehicular edge computing (VEC) has generated enormous attention in recent years. It pushes the computational resources in close proximity to the data sources and thus caters for the explosive growth of vehicular applications. Owing to the high mobility of vehicles, these applications are of latency-sensitive requirements in most cases. Accordingly, such requirements still pose a great challenge to the computing capabilities of VEC, when these applications are outsourced and executed in VEC. Against this backdrop, we propose a new mathematical model which respectively generalizes the computation and communication models, and applies application oriented caching into VEC in this paper. Based on this model, a new strategy is further proposed to optimize the average response time of applications over an infinite time-slotted horizon for VEC. A long-term energy consumption constraint is imposed to guarantee the stability of VEC system, and the Lyapunov optimization technology is adopted to tackle this constraint issue. Two greedy heuristics are put forward to help find the approximate optimal solution in the drift-plus-penalty based algorithm. Extensive experiments have been conducted to evaluate the response time and energy consumption in the caching assisted VEC. The simulation results have shown that the proposed strategy can dramatically optimize the average response time while satisfying the long-term energy consumption constraint.

With the rapid development of the Internet of Things (IoT) and communication technology, Deep Neural Network (DNN) applications like computer vision, can now be widely used in IoT devices. However, due to the insufficient memory, low computing capacity, and low battery capacity of IoT devices, it is difficult to support the high-efficiency DNN inference and meet users' requirements for Quality of Service (QoS). Worse still, offloading failures may occur during the massive DNN data transmission due to the intermittent wireless connectivity between IoT devices and the cloud. In order to fill this gap, we consider the partitioning and offloading of the DNN model, and design a novel optimization method for parallel offloading of large-scale DNN models in a local-edge-cloud collaborative environment with limited resources. Combined with the coupling coordination degree and node balance degree, an improved Double Dueling Prioritized deep Q-Network (DDPQN) algorithm is proposed to obtain the DNN offloading strategy. Compared with existing algorithms, the DDPQN algorithm can obtain an efficient DNN offloading strategy with low delay, low energy consumption, and low cost under the premise of ensuring ''delay-energy-cost'' coordination and reasonable allocation of computing resources in a local-edge-cloud collaborative environment.

With the popularity of mobile devices, intelligent applications, e.g., face recognition, intelligent voice assistant, and gesture recognition, have been widely used in our daily lives. However, due to the lack of computing capacities, it is difficult for mobile devices to support complex Deep Neural Network (DNN) inference. To alleviate the pressure on these devices, traditional methods usually upload part of the DNN model to a cloud server and perform a DNN query after uploading an entire DNN model. To achieve real-time DNN query, we consider the collaboration between local, edge and cloud, and perform DNN query when uploading DNN partitions. In this paper, we propose an Efficient offloading scheme for DNN Inference Acceleration (EosDNN) in a local-edge-cloud collaborative environment, where the DNN inference acceleration is mainly embodied in the optimization of migration delay and realization of real-time DNN query. EosDNN comprehensively considers the migration plan and uploading plan, where for the former, a Particle Swarm Optimization with Genetic Algorithm (PSO-GA) is applied to obtain the distribution of DNN layers under the server with the lowest migration delay, and for the latter, a Layer Merge Uploading Algorithm (LMU) is proposed to obtain DNN partitions and their upload order with efficient DNN query performance. Experimental results demonstrate that EosDNN can be applied to large-scale DNN model migration, which can achieve an ideal migration delay and obtain a more fine-grained DNN partition uploading plan, thereby optimizing DNN query performance.

A variety of methods have been proposed for modeling and mining dynamic complex networks, in which the topological structure varies with time. As the most popular and successful network model, the stochastic block model (SBM) has been extended and applied to community detection, link prediction, anomaly detection, and evolution analysis of dynamic networks. However, all current models based on the SBM for modeling dynamic networks are designed at the community level, assuming that nodes in each community have the same dynamic behavior, which usually results in poor performance on temporal community detection and loses the modeling of node abnormal behavior. To solve the above-mentioned problem, this article proposes a hierarchical Bayesian dynamic SBM (HB-DSBM) for modeling the node-level and community-level dynamic behavior in a dynamic network synchronously. Based on the SBM, we introduce a hierarchical Dirichlet generative mechanism to associate the global community evolution with the microscopic transition behavior of nodes near-perfectly and generate the observed links across the dynamic networks. Meanwhile, an effective variational inference algorithm is developed and we can easy to infer the communities and dynamic behaviors of the nodes. Furthermore, with the two-level evolution behaviors, it can identify nodes or communities with abnormal behavior. Experiments on simulated and real-world networks demonstrate that HB-DSBM has achieved state-of-the-art performance on community detection and evolution. In addition, abnormal evolutionary behavior and events on dynamic networks can be effectively identified by our model.

Advances in information and communication technologies have significantly influenced the operation of low-voltage distribution grids. As essential elements of distribution grids, user-side smart meters find many smart grid applications, for example to measure electrical energy use and facilitate communications. However, the service models of distribution grids remain under development in association with upgrading of user-side smart meters. These meters are resource constrained, and challenging to upgrade on a large scale. To address this issue, this article describes a container-driven service architecture, in which containers are used to create a virtual dedicated agent (digital twin) for each user-side smart meter. The agent can be deployed either in the cloud or on an edge system, and can be upgraded to support emerging smart grid applications, thus minimizing the future upgrading requirements of user-side smart meters. We built experimental test beds to verify the proposed architecture and evaluated its performance in real-world experiments.

The recent surge in the number of connected vehicles and vehicular applications really benefits citizens. Various vehicular applications are developed to cater for the increasingly sophisticated demands of drivers. Against this background, vehicular edge computing (VEC) is put forward as a promising solution to meet the strict latency requirement of these vehicular applications, by undertaking the computation offloaded from the nearby vehicles. Furthermore, task-oriented caching strategies are also applied to VEC for performance improvement. However, challenges faced by caching-enabled VEC still need to be addressed. For example, many factors can restrict the application of task caching in VEC, which usually include limited caching capability, extra energy consumption incurred by task caching, caching results delivery and so on. To overcome these issues, we propose a general caching-enabled VEC scheme and aim to jointly optimize the task caching and computation offloading in the VEC system. Moreover, we consider not only the response latency reduction benefitting from task caching, but also the energy consumption incurred by task caching. In particular, we strive to minimize the weighted sum of the service time and energy consumption for all the offloading requests in VEC. Due to the exponential time taken to obtain the optimal value, we in this paper propose a genetic algorithm-based task caching and computation offloading strategy. Extensive simulation has been carried out to investigate its efficiency compared to the benchmark algorithms. The simulation results reveal that the proposed strategy outperforms other approaches including the greedy approach and the random approach.

Deoxyribonucleic acid (DNA) is an attractive medium for long-term digital data storage due to its extremely high storage density, low maintenance cost and longevity. However, during the process of synthesis, amplification and sequencing of DNA sequences with homopolymers of large run-length, three different types of errors, namely, insertion, deletion and substitution errors frequently occur. Meanwhile, DNA sequences with large imbalances between GC and AT content exhibit high dropout rates and are prone to errors. These limitations severely hinder the widespread use of DNA-based data storage. In order to reduce and correct these errors in DNA storage, this paper proposes a novel coding schema called DNA-LC, which converts binary sequences into DNA base sequences that satisfy both the GC balance and run-length constraints. Furthermore, our coding mode is able to detect and correct multiple errors with a higher error correction capability than the other methods targeting single error correction within a single strand. The decoding algorithm has been implemented in practice. Simulation results indicate that our proposed coding scheme can offer outstanding error protection to DNA sequences. The source code is freely accessible at https://github.com/XiayangLi2301/DNA.

Identifying high-order Single Nucleotide Polymorphism (SNP) interactions of additive genetic model is crucial for detecting complex disease gene-type and predicting pathogenic genes of various disorders. We present a novel framework for high-order gene interactions detection, not directly identifying individual site, but based on Deep Learning (DL) method with Differential Privacy (DP), termed as Deep-DPGI. Firstly, integrate loss functions including cross-entropy and focal loss function to train the model parameters that minimize the value of loss. Secondly, use the layer-wise relevance analysis method to measure relevance difference between neurons weight and outputting results. Deep-DPGI disturbs neuron weight by adaptive noising mechanism, protecting the safety of high-order gene interactions and balancing the privacy and utility. Specifically, more noise is added to gradients of neurons that is less relevance with the outputs, less noise to gradients that more relevance. Finally, Experiments on simulated and real datasets demonstrate that Deep-DPGI not only improve the power of high-order gene interactions detection in with marginal and without marginal effect of complex disease models, but also prevent the disclosure of sensitive information effectively.

Network embedding has been successfully used for a variety of tasks, e.g., node clustering, community detection, link prediction and evolution analysis on complex networks. For a given network, embedding methods are usually designed based on first-order proximity, second-order proximity, community constraints, etc. However, they are incapable of capturing the structural similarity of nodes. The bridge nodes with small proximity and located in different communities, should be similar in embedding space since they have the same surrounding structure. In this paper, these structural features are referred to as lower-order information, which could reveal and modify the structural similarity of nodes in the embedding space. Specifically, we propose to construct the feature matrix with the lower-order information of the network. In order to effectively fuse the structural features of nodes into embedding space, an intuitive, interpretable and feasible method named LONE-NMF is proposed, which adopts the representation learning framework based on non-negative matrix factorization. It can effectively learn the representation vectors of nodes in the network via preserving the proximity and lower-order information. Moreover, an optimization algorithm is designed for LONE-NMF. Extensive experiments based on clustering and link prediction show that the proposed method achieves significant performance improvement comparing with some baselines. Finally, we validate the principle and advantage of LONE-NMF through a case study.

Two functions are essential and necessary for the wireless-powered communication network, which are energy beamforming and localization. On one hand, energy beamforming controls the wireless energy waves of the energy access point (E-AP) in order to activate the nodes for transmitting information. On the other hand, locating the nodes is important to network management and location-based services in the wireless power communication network (WPCN). For a large-scale network, cooperative localization that employs neighborhood nodes to participate in positioning unknown target nodes is highly accurate and efficient. However, how to use energy beamforming to achieve highly accurate localization is not fully investigated yet. In this article, we analyze the impacts of energy beamforming on the cooperative localization performance of WPCNs. We formulate the Fisher information matrix (FIM) and the corresponding Cramer-Rao lower bound (CRLB) for the full connected network and a single node, respectively. Then, we propose beamforming schemes to optimize the cooperative localization and the power consumption. For optimal localization problems, we derive the closed-form expression of the optimal energy beamforming. For the optimal energy efficiency problems, we propose semidefinite programming (SDP) solutions to achieve the minimum power consumption while using calibrations to approach the actual localization requirements. Further, we also analyze the impacts of channel uncertainty. Through extensive simulations, the results demonstrate the dominant factors of the localization performance, and the performance improvements of our proposed schemes, which outperform the existing power allocation schemes.

Great achievements have been made in network embedding based on single-layer networks. However, there are a variety of scenarios and systems that can be presented as multiplex networks, which can reveal more interesting patterns hidden in the data compared to single-layer networks. In the field of network embedding, in order to project the multiplex network into the latent space, it is necessary to consider richer structural information among network layers. However, current methods for multiplex network embedding mostly focus on the similarity of nodes in each layer of the network, while ignoring the similarity between different layers. In this paper, for multiplex network embedding, we propose a Layer Information Similarity Concerned Network Embedding (LISCNE) model considering the similarities between layers. Firstly, we introduce the common vector for each node shared by all layers and layer vectors for each layer where common vectors obtain the overall structure of the multiplex network and layer vectors learn semantics for each layer. We get the node embeddings in each layer by concatenating the common vectors and layer vectors with the consideration that the node embedding is related not only to the surrounding neighbors but also to the overall semantics. Furthermore, we define an index to formalize the similarity between different layers and the cross-network association. Constrained by layer similarity, the layer vectors with greater similarity are closer to each other and the aligned node embedding in these layers is also closer. To evaluate our proposed model, we conduct node classification and link prediction tasks to verify the effectiveness of our model, and the results show that LISCNE can achieve better or comparable performance compared to existing baseline methods.

As a key technology in the 5G era, Mobile Edge Computing (MEC) has developed rapidly in recent years. MEC aims to reduce the service delay of mobile users, while alleviating the processing pressure on the core network. MEC can be regarded as an extension of cloud computing on the user side, which can deploy edge servers and bring computing resources closer to mobile users, and provide more efficient interactions. However, due to the user's dynamic mobility, the distance between the user and the edge server will change dynamically, which may cause fluctuations in Quality of Service (QoS). Therefore, when a mobile user moves in the MEC environment, certain approaches are needed to schedule services deployed on the edge server to ensure the user experience. In this paper, we model service scheduling in MEC scenarios and propose a delay-aware and mobility-aware service management approach based on concise probabilistic methods. This approach has low computational complexity and can effectively reduce service delay and migration costs. Furthermore, we conduct experiments by utilizing multiple realistic datasets and use iFogSim to evaluate the performance of the algorithm. The results show that our proposed approach can optimize the performance on service delay, with 8% to 20% improvement and reduce the migration cost by more than 75% compared with baselines during the rush hours.

Power conversion efficiency (PCE) of SnO2-based dye-sensitized solar cells (DSSCs) is reported to increase slowly over a period of days after their fabrication. Thus far, underlying processes involved in this slowly increasing phenomenon of PCE are not understood. In this article, we investigate the spontaneous enhancement of PCE in SnO2-based DSSCs and the underlying mechanism by using the X-ray diffraction, the X-ray photoelectron spectra (XPS), electrochemical impedance spectrum, and first-principles calculations. The Sn 3d and O 1s binding energy are observed to shift toward low binding energy over time by the XPS measurements, and the ratio of O/Sn on the surface of SnO2 increases from 1.9 to 2.1 implying the adsorption of oxygen atom on the surface of SnO2. Our first-principles calculation results confirm that a shift-up in the conduction band minimum of SnO2 owing to its adsorption of oxygen atom. We, therefore, attribute the slowly increasing phenomenon of PCE to oxygen adsorption effect on the surface of SnO2. This article gives a comprehensive understanding of the surface oxidation effect on the PCE enhancement in DSSCs and provides an effective way to fabricate DSSCs based on oxide semiconductors with good performance and long-time stability.

Fog/Edge computing emerges as a novel computing paradigm that harnesses resources in the proximity of the Internet of Things (IoT) devices so that, alongside with the cloud servers, provide services in a timely manner. However, due to the ever-increasing growth of IoT devices with resource-hungry applications, fog/edge servers with limited resources cannot efficiently satisfy the requirements of the IoT applications. Therefore, the application placement in the fog/edge computing environment, in which several distributed fog/edge servers and centralized cloud servers are available, is a challenging issue. In this article, we propose a weighted cost model to minimize the execution time and energy consumption of IoT applications, in a computing environment with multiple IoT devices, multiple fog/edge servers and cloud servers. Besides, a new application placement technique based on the Memetic Algorithm is proposed to make batch application placement decision for concurrent IoT applications. Due to the heterogeneity of IoT applications, we also propose a lightweight pre-scheduling algorithm to maximize the number of parallel tasks for the concurrent execution. The performance results demonstrate that our technique significantly improves the weighted cost of IoT applications up to 65 percent in comparison to its counterparts.

Among the novel IT paradigms, cloud computing and the Internet of Things (CloudIoT) are two complementary areas designed to support the creation of smart cities and application services. The CloudIoT not only presents ubiquitous services through IoT nodes but it also provides virtually unlimited resources through services composition. The services composition problem aims to find a set of services among functionally equivalent services with different Quality of Service (QoS) concerning users’ constraints. To this aim, previous studies calculate QoS values through service logs without considering the presence of anomalies in the existing QoS values; however, the dynamicity of distributed service environments and communication networks in CloudIoT environments causes anomalies in the QoS values. Therefore, existing approaches fail to model QoS values accurately that leads to service-level agreement (SLA) violation and penalties for service broker. To address this challenge, we propose a scalable anomaly-aware approach (SAIoT) including two main components: the first component models QoS values based on a machine learning anomaly detection technique, to remove the existing abnormal QoS records, and the second component finds a near-optimal composition by using an effective and efficient metaheuristic algorithm. The experimental results based on real-world data sets show that our approach achieves 30.64% of the average improvement in the QoS value of a composite plan with equal or even less price compared to the previous works, such as information theory-based and advertised QoS-based methods.

With the proliferation of compute-intensive and delay-sensitive mobile applications, large amounts of computational resources with stringent latency requirements are required on Internet-of-Things (IoT) devices. One promising solution is to offload complex computing tasks from IoT devices either to mobile-edge computing (MEC) or mobile cloud computing (MCC) servers. MEC servers are much closer to IoT devices and thus have lower latency, while MCC servers can provide flexible and scalable computing capability to support complicated applications. To address the tradeoff between limited computing capacity and high latency, and meanwhile, ensure the data integrity during the offloading process, we consider a blockchain scenario where edge computing and cloud computing can collaborate toward secure task offloading. We further propose a blockchain-enabled IoT-Edge-Cloud computing architecture that benefits both from MCC and MEC, where MEC servers offer lower latency computing services, while MCC servers provide stronger computation power. Moreover, we develop an energy-efficient dynamic task offloading (EEDTO) algorithm by choosing the optimal computing place in an online way, either on the IoT device, the MEC server or the MCC server with the goal of jointly minimizing the energy consumption and task response time. The Lyapunov optimization technique is applied to control computation and communication costs incurred by different types of applications and the dynamic changes of wireless environments. During the optimization, the best computing location for each task is chosen adaptively without requiring future system information as prior knowledge. Compared with previous offloading schemes with/without MEC and MCC cooperation, EEDTO can achieve energy-efficient offloading decisions with relatively lower computational complexity.

Channel state information (CSI) feedback plays an important part in frequency division duplex (FDD) massive multiple-input multiple-output (MIMO) systems. However, it is still facing many challenges, e.g., excessive feedback overhead, low feedback accuracy and a large number of training parameters. In this letter, to address these practical concerns, we propose a deep learning (DL)-based CSI feedback scheme, named DS-NLCsiNet. By taking advantage of non-local blocks, DS-NLCsiNet can capture long-range dependencies efficiently. In addition, dense connectivity is adopted to strengthen the feature refinement module. Simulation results demonstrate that DS-NLCsiNet achieves higher CSI feedback accuracy and better reconstruction quality for the same compression ratio, when compared to state-of-the-art compression schemes.

Tin oxide (SnO2) film is a promising photoanode for dye-sensitized solar cells (DSCs) with excellent optical and electrical properties as well as facile accessibility. Highly efficient DSCs require a fast charge transfer from dye to SnO2 in order to reduce the recombination from trap states. Herein, we report a Cu-doped SnO2 film as a photoanode for DSCs with faster charge transfer and higher efficiency. Ultraviolet photoelectron spectroscopy study indicates that conduction band of the Cu-doped SnO2 is adjusted from − 4.75 to − 4.57 eV via 5 at.% Cu concentrations. By adjusting the Cu concentrations, energy offset between the SnO2 and dye is reduced, which accelerates the charge transfer. Power conversion efficiency of the DSCs with optimized Cu-doped SnO2 achieves 4.01%, which is 61.7% higher than its counterpart without Cu doping.

City Internet-of-Things (IoT) applications are becoming increasingly complicated and thus require large amounts of computational resources and strict latency requirements. Mobile cloud computing (MCC) is an effective way to alleviate the limitation of computation capacity by offloading complex tasks from mobile devices (MDs) to central clouds. Besides, mobile-edge computing (MEC) is a promising technology to reduce latency during data transmission and save energy by providing services in a timely manner. However, it is still difficult to solve the task offloading challenges in heterogeneous cloud computing environments, where edge clouds and central clouds work collaboratively to satisfy the requirements of city IoT applications. In this article, we consider the heterogeneity of edge and central cloud servers in the offloading destination selection. To jointly optimize the system utility and the bandwidth allocation for each MD, we establish a hybrid offloading model, including the collaboration of MCC and MEC. A distributed deep learning-driven task offloading (DDTO) algorithm is proposed to generate near-optimal offloading decisions over the MDs, edge cloud server, and central cloud server. Experimental results demonstrate the accuracy of the DDTO algorithm, which can effectively and efficiently generate near-optimal offloading decisions in the edge and cloud computing environments. Furthermore, it achieves high performance and greatly reduces the computational complexity when compared with other offloading schemes that neglect the collaboration of heterogeneous clouds. More precisely, the DDTO scheme can improve computational performance by 63%, compared with the local-only scheme.

The buoyancy of Microcystis colonies determines the occurrence and dominance of bloom on the water surface. Besides the cell density regulation and the formation of larger size aggregates, increases in cell volume per colony (Vcell) and the colony's compactness (i.e., volume ratio of cells to the colony, VR) may promote Microcystis colony buoyancy. Yet only a few studies have studied the relationship between the internal structure variation of colonies and their buoyancy, and the co-regulatory role of Vcell and VR of Microcystis colonies in the floating velocity (FV) remains largely unexplored. In the present study, we optimized a method for measuring the compactness of Microcystis colonies based on the linear relationship between total Vcell and chlorophyll a. Different relationships between the VRs and FVs were observed with different colony size and Vcell range groups. Both field and laboratory experiments showed that FV/(D50, median diameter)2 had a significant linear relationship with VR, indicating that the cell density and extracellular polysaccharides were unchanged over a short time period and could be estimated via the slope and intercept of a fitted line. We also constructed a functional relationship between FV, VR, and Vcell and found that high VR and Vcell can promote Microcystis buoyancy. This means that increasing cell compactness or Vcell may be an active regulation strategy for Microcystis colonies to promote buoyancy. Therefore, quantifying the internal structure of Microcystis colonies is strongly recommended for the assessment of Microcystis bloom development and their management. Graphical abstract.

Community detection and community evolution tracking are two important tasks in dynamic complex network analysis. Recently, a variety of models and methods have been proposed for detecting the community structure and analyzing their evolution. However, all these methods are only committed to improving the performance of community detection or identifying evolutionary events, ignoring the internal relevance between the structure of each snapshot of the dynamic network and the evolution pattern of communities, especially the structural features of nodes and their dynamic transition behavior. To cope with this problem, we firstly conduct experiments on 15 real-world dynamic networks to explore the transition behavior of nodes in dynamic networks, which is one of the most influential evolutionary patterns in temporal community detection. Firstly, we obtain the temporal community structure based on very successful temporal community detection methods. Secondly, we extract features of nodes based on the structure of the dynamic network, and take the community transition behavior of nodes as the binary classification problem. Finally, we use the decision tree to find the node-level features that have a general impact on node transition. Experiments indicate that the degree and average neighbor degree of nodes have the most common indispensable impact on the node transition behavior, which are very helpful for modeling dynamic complex networks in future.

As a powerful tool for machine learning on the graph, network embedding, which projects nodes into low-dimensional spaces, has a variety of applications on complex networks. Most current methods and models are not suitable for bipartite networks, which have two different types of nodes and there are no links between nodes of the same type. Furthermore, the only existing methods for bipartite network embedding ignore the internal mechanism and highly nonlinear structures of links. Therefore, in this paper, we propose a new deep learning method to learn the node embedding for bipartite networks based on the widely used autoencoder framework. Moreover, we carefully devise a node-level triplet including two types of nodes to assign the embedding by integrating the local and global structures. Meanwhile, we apply the variational autoencoder (VAE), a deep generation model with natural advantages in data generation and reconstruction, to enhance the node embedding for the highly nonlinear relationships between nodes and complex features. Experiments on some widely used datasets show the effectiveness of the proposed model and corresponding algorithm compared with some baseline network (and bipartite) embedding techniques.

Mobile cloud offloading migrates heavy computation from mobile devices to remote cloud resources or nearby cloudlets. It is a promising method to alleviate the struggle between resource-constrained mobile devices and resource-hungry mobile applications. Caused by frequently changing location mobile users often see dynamically changing network conditions which have a great impact on the perceived application performance. Therefore, making high-quality offloading decisions at run time is difficult in mobile environments. To balance the energy-delay tradeoff based on different offloading-decision criteria (e.g., minimum response time or energy consumption), an energy-efficient offloading-decision algorithm based on Lyapunov optimization is proposed. The algorithm determines when to run the application locally, when to forward it directly for remote execution to a cloud infrastructure and when to delegate it via a nearby cloudlet to the cloud. The algorithm is able to minimize the average energy consumption on the mobile device while ensuring that the average response time satisfies a given time constraint. Moreover, compared to local and remote execution, the Lyapunov-based algorithm can significantly reduce the energy consumption while only sacrificing a small portion of response time. Furthermore, it optimizes energy better and has less computational complexity than the Lagrange Relaxation based Aggregated Cost (LARAC-based) algorithm.

Deoxyribonucleic acid (DNA)-based data storage is a promising new storage technology which has the advantage of high storage capacity and long storage time compared with traditional storage media. However, the synthesis and sequencing process of DNA can randomly generate many types of errors, which makes it more difficult to cluster DNA sequences to recover DNA information. Currently, the available DNA clustering algorithms are targeted at DNA sequences in the biological domain, which not only cannot adapt to the characteristics of sequences in DNA storage, but also tend to be unacceptably time-consuming for billions of DNA sequences in DNA storage. In this paper, we propose an efficient DNA clustering method termed Clover for DNA storage with linear computational complexity and low memory. Clover avoids the computation of the Levenshtein distance by using a tree structure for interval-specific retrieval. We argue through theoretical proofs that Clover has standard linear computational complexity, low space complexity, etc. Experiments show that our method can cluster 10 million DNA sequences into 50 000 classes in 10 s and meet an accuracy rate of over 99%. Furthermore, we have successfully completed an unprecedented clustering of 10 billion DNA data on a single home computer and the time consumption still satisfies the linear relationship. Clover is freely available at https://github.com/Guanjinqu/Clover.