As a promising distributed computing paradigm, edge computing (EC) enhances service quality and reduces service latency by deploying computational and storage resources at the network edge. However, due to edge servers' geographic distribution and resource constraints, EC is challenged by edge denial-of-service (EDDoS) attacks. Although various approaches have been proposed to mitigate EDDoS attacks, the impact of capacity uncertainties in edge server processing capacity and transmission latency has been largely overlooked. This limits the adaptability and robustness of mitigation strategies in edge computing environments, which could result in overload and paralysis of edge servers. To address this issue, this paper introduces uEDDoS-D, an uncertainty-aware EDDoS mitigation approach based on an improved deep deterministic policy gradient algorithm. uEDDoS-D models the uncertainties of computing capacities and transmission latency of edge servers in EC environments and leverages the collective computational resources of edge servers to mitigate EDDoS attacks without relying on traditional attack detection mechanisms. uEDDoS-D aims to minimize the impact of uncertainties in edge server processing capacity and transmission latency on mitigation decisions while effectively reducing service latency. Experimental results demonstrate that uEDDoS-D significantly reduces service latency, outperforming the state-of-the-art approaches by an average of 16.75%, and enhances system robustness in EDDoS attack scenarios.

Multivariate time series anomaly detection is crucial for the reliability and stability of operations in cloud-edge computing systems. However, fast model training requirements, unlabeled datasets, and high-dimensional time series made it challenging to build a model that can quickly and accurately localize anomalies with good robustness. Therefore, this paper proposes a robust anomaly detection approach with spatiotemporal representation learning for multivariate time series (RAD-SRL) to address the above challenges. RAD-SRL leverages improved Dilated Causal Convolutions (DCNs) to extract spatiotemporal representations in MTS and incorporates a multi-head self-attention mechanism and an Update Gate Recurrent Neural Network (UGRNN) to capture multiple-dimensional information. Moreover, the self-adjusting mechanism of RAD-SRL reduces computational resource consumption and improves the generalizability of the model. Experimental studies on four public datasets demonstrate that the RAD-SRL outperforms the baseline methods in terms of both detection performance and time efficiency.

Recent latent diffusion models (LDMs) have been explored to generate diverse domain-specific images based on source domain data, showing promising performance in domain generalization tasks. However, although the generated images present counterfactual augmentation, such as the background and style changes, the distortion of object details disrupts the causal factors, such as texture and shape. This leads to negative outcomes when directly applying LDM to domain generalization in object detection. To address the problems mentioned above, we propose Object-Preserving Counterfactual Diffusion augmentation method (OPCD) to explore the diffusion model to generate diverse domain-specific images without disrupting the object details. First, we construct a region-aware image generation framework, which leverages labeled source domain data to guide LDM in generating region-constrained images that preserve the semantic consistency of the original source images. Second, we propose object-preserving counterfactual augmentation, which retains the object region of the generated image and fuses diversified global information. This ensures that object details are not distorted and that the generated information is maintained. Third, to reduce the resource burden of generating a large number of images in LDM, we design a random insertion strategy. It mixes generated and source domain images, turning limited diversity samples into abundant training data. Experimental results on several benchmark datasets show that OPCD outperforms existing methods in single-domain generalized object detection. Codes can be found at https://github.com/qinhongda8/OPCD.

In the industrial Internet, intelligent operation and maintenance can ensure the secure and stable operation of industrial software. To achieve effective intelligent operation and maintenance, it’s essential to conduct time series anomaly detection within software systems and other devices. To enhance the detection effectiveness, extensive research has been conducted. However, multivariate time series are composed of high-dimensional, high-noise, and random data. These anomalies are both subtle and dense. It is difficult to detect anomalies accurately. The increased complexity of existing methods results in lower operational efficiency. To address these issues, this paper proposes the time series anomaly detection method DG-LSFNet. DG-LSFNet adeptly captures feature correlations across various temporal states to extract additional valuable insights. Next, DG-LSFNet establishes long-term and short-term temporal correlations. It can capture normal information within short intervals between anomalies, thereby reducing false alarm in anomaly detection. Then, DG-LSFNet approximate and unearth the original data information to improve anomaly detection performance. In addition, DG-LSFNet reduces time complexity by simplifying the model structure and effectively improves the efficiency of anomaly detection. This paper conducts experiments to compare DG-LSFNet with state-of-the-art methods on four benchmark datasets. The experimental results indicate that DG-LSFNet outperforms state-of-the-art methods in terms of anomaly detection performance, enhancing the interpretability of anomaly detection.

Industrial part surface defect detection aims to precisely locate defects in images, which is crucial for quality control in manufacturing. The traditional method needs to be designed in advance, but it has shortcomings in terms of generalization ability. Deep neural network-based defect detection faces issues such as low resolution, data scarcity, labeling costs, and high computation. Therefore, an improved solution is necessary to enhance industrial product quality control efficiency and accuracy. A new method is proposed that utilizes end-to-end training of a two-stage neural network based on segmentation with an extended training process. The gradient flow from the classification to the segmentation network is adjusted to prevent unstable features from interfering with learning. To address the problem of image oversampling and undersampling during training, a frequency sampling scheme for negative samples is introduced. Additionally, positive pixels in the region-based segmentation mask are weighted using the distance transform algorithm so that regions with a high probability of defects can be detected without detailed annotation. Experiments are conducted across three distinct defect datasets by applying hybrid supervision encompassing diverse conditions. In the optimal case, the AP rate of the proposed model on the DAGM and KolektorSDD datasets reaches 100%, and the AP rate on the Severstal Steel dataset reaches 98.99%. The experiments’ results show that the detection accuracy has improved greatly, showing how well the proposed method works in the industry.

A safety boundary wall in VR is a virtual barrier that defines a safe area, allowing users to navigate and interact without safety concerns. However, existing implementations neglect to utilize the safety boundary wall's large surface for displaying interactive information. In this work, we propose the BoundaryScreen technique based on the "walking outward" metaphor to add interactivity to the safety boundary wall. Specifically, we augment the safety boundary wall by placing the home screen on it. To summon the home screen, the user only needs to walk outward until it appears. Results showed that (i) participants significantly preferred BoundaryScreen in the outermost two-step-wide ring-shaped section of a circular safety area; and (ii) participants exhibited strong "behavioral inertia" for walking, i.e., after completing a routine activity involving constant walking, participants significantly preferred to use the walking-based BoundaryScreen technique to summon the home screen.

Under the background of complicated fault detection scenarios and diversified data in the current power distribution grid, it is a significant challenge to deploy high-performance fault detection models for efficient multi-task collaboration processing on lightweight edge intelligent devices. A collaborative multi-teacher distillation for multi-task fault detection in the power distribution grid(CMT-KD) is introduced, which enhances multi-task fault detection by redefining Residual Network (ResNet). We further implement model compression and acceleration by adopting the weight quantization method, while enhancing the performance of the multi-task fault detector via dynamically adapting the knowledge weights from multiple teachers. The proposed method allows for the detection of multiple power distribution grid faults with a single model, significantly reducing the number of model parameters, floating point operations, and training time, while also demonstrating favorable performance in terms of improving accuracy and reducing prediction errors.

With the increasing performance of deep convolutional neural networks, they have been widely used in many computer vision tasks. However, a huge convolutional neural network model requires a lot of memory and computing resources, which makes it difficult to meet the requirements of low latency and reliability of edge computing when the model is deployed locally on resource-limited devices in edge environments. Quantization is a kind of model compression technology, which can effectively reduce model size, calculation cost and inference delay, but the quantization noise will cause the accuracy of the quantization model to decrease. Aiming at the problem of precision loss caused by model quantization, this paper proposes a post-training quantization method based on scale optimization. By reducing the influence of redundant parameters in the model on the quantization parameters in the process of model quantization, the scale factor optimization is realized to reduce the quantization error and thus improve the accuracy of the quantized model, reduce the inference delay and improve the reliability of edge applications. The experimental results show that under different quantization strategies and different quantization bit widths, the proposed method can improve the accuracy of the quantized model, and the absolute accuracy of the optimal quantization model is improved by 1.36%. The improvement effect is obvious, which is conducive to the application of deep neural network in edge environment.

Log anomaly detection on edge devices is one of the key components to ensure system security and diagnose system faults. Existing log anomaly detection methods usually contain a large number of parameters and computations. However, processing large-scale logs on edge devices is still a bottleneck due to the limited computational power of edge devices, which can hardly afford the complexity of these detection models. Hence, we propose a lightweight and efficient log anomaly detection method called TinyLog, which aims to facilitate the identification of anomalous log patterns in large-scale IoT systems. Specifically, TinyLog uses Word2Vec and TF-IDF to construct weighted log semantic vector. Further, a lightweight temporal convolutional network based on depthwise separable convolution and triplet attention is designed for anomaly detection in edge devices. Experimental results show that TinyLog outperforms other state-of-the-art benchmark methods in terms of performance and efficiency, while its computational overhead is extremely low. Thus, our approach can be effectively applied to resource-constrained edge devices.

In the Industrial Internet of Things (IIoT), accurate time series anomaly detection is key to ensuring operational efficiency and safety. The current challenge is to process and analyze the high-dimensional data collected over the long term, especially the combined temporal and spatial dependence, which are crucial to the accuracy of anomaly detection. The existing methods often overlook the spatio-temporal dual dependence of time series data, resulting in inevitable false positives. This paper proposes an anomaly detection model capturing long-term spatio-temporal dependence for multi-dimensional time series (LSTD-MTS) to solve the above problems. LSTD-MTS utilizes Multi-Scale Temporal Convolutional Network (MSTCN) and Temporal-GRU components to enhance the model's ability to capture temporal dependence of long-term series, combines relational vector graph learning and Graph Attention Network (GAT) to capture multi-dimensional spatial dependence, and maintains the linear relationship of data through autoregressive (AR) component. Compared with seven baseline methods across five datasets, LSTD-MTS demonstrates superior detection performance and greater robustness than those of the baseline methods. Its F1 score is at least 2.76% higher than baseline methods, which supports the robust long-term multi-dimensional analysis ability of LSTD-MTS.

Log parsing is the initial stage of automated log analysis, which mainly involves converting semi-structured log messages into structured log templates. However, the initial log data size of software systems and services is relatively small, which limits the effectiveness of existing $\log$ parsers. Traditional data-driven log parsers use the structured nature of logs to construct domain rules, which may make it difficult to maintain stable performance when dealing with log data of different sizes and types due to mismatches. To address the limitations of existing methods, this paper proposes a log parsing method called LogSPL, which captures high-quality samples from small-scale log data based on semantics and extracts log templates using prompt learning. The research in this paper performs extensive experiments on 16 benchmark $\log$ datasets. The results show that LogSPL improves parsing accuracy and group accuracy by 3.2% and 2.6% over the optimal parser.

With the support of 5G technology, mobile edge computing has made the application of industrial IoT and power IoT more and more extensive. By deploying a certain number of edge servers at the edge of the network, network service delay may significantly reduce. For the IoT scenario where the content demand is unpredictable, there are multiple distributed cloud servers and the distributed cloud servers do not communicate directly, a feasible way to improve the network service quality is to dynamically optimize the storage of edge servers and formulate targeted caching strategies. This paper proposes an edge caching approach based on multi-agent deep deterministic policy gradient named MADDPG-C, which regards distributed cloud servers and edge servers as different types of agents and maximizes the efficiency of edge caching in cooperation and competition. Simulation experiments show that the proposed MADDPG-C can further improve the hit rate of the edge cache and reduce the waiting delay of terminal devices.

The pre-training language model BERT has brought significant performance improvements to a series of natural language processing tasks, but due to the large scale of the model, it is difficult to be applied in many practical application scenarios. With the continuous development of edge computing, deploying the models on resource-constrained edge devices has become a trend. Considering the distributed edge environment, how to take into account issues such as data distribution differences, labeling costs, and privacy while the model is shrinking is a critical task. The paper proposes a new BERT distillation method with source-free unsupervised domain adaptation. By combining source-free unsupervised domain adaptation and knowledge distillation for optimization and improvement, the performance of the BERT model is improved in the case of cross-domain data. Compared with other methods, our method can improve the average prediction accuracy by up to around 4% through the experimental evaluation of the cross-domain sentiment analysis task.

The rapid development of industrial IoT communication and the wide application of smart terminal devices, large-scale mobile data traffic and frequent data requests pose challenges to resource-constrained wireless communication networks. In order to reduce the network load, the introduction of content encoding caching to the network edge is considered an effective solution. Due to the heterogeneity of the resources of the mobile edge system, the variability of user requests, and the mobility of users, the coded caching strategy at the mobile edge needs to be continuously optimized. However, existing articles do not adequately investigate how to intelligently update the coded caching policy in dynamic environments. For this reason, this paper proposes a heterogeneous multi-agent MDS coded caching scheme (MACC) based on value decomposition, which considers the edge caching server with storage capacity and the mobile user as two different types of agents and improves the caching policy through interaction. In addition, considering the heterogeneous preference of heterogeneous multi-agent and the fact that an increase in the number may lead to a sharp increase in the training dimension, the idea of a value decomposition network is introduced in the training learning of the coded caching scheme. Simulation experiments verify that the proposed MACC can effectively reduce the load on the forward link and achieve a higher cache hit rate.

In mobile edge computing, DNN-driven intelligent inference service is highly sensitive to latency.Recently, collaborative inference between user devices and Edge Servers (ESs) based on DNN partition has been used in service acceleration.However, due to the limited computing resources of ESs, there is resource competition between concurrent requests, resulting in the partition tasks cannot be offloaded to ESs in time.Therefore, it is necessary to design an efficient offloading scheme for partitionbased concurrent inference tasks.Existing task offloading schemes based on Deep Reinforcement Learning (DRL) can solve complex decision-making problems in high-dimensional state space, but there are problems such as insufficient sample diversity and easily falling into local optimum.Therefore, we propose a collaborative DNN inference task offloading scheme based on distributed Soft Actor-Critic(SAC).It supports SAC Agents to explore samples in parallel and share learning experiences, and improves the randomness of the policy through the maximum entropy mechanism to avoid falling into local optimum, thus achieving efficient offloading of concurrent partition tasks.Experimental results on DNN benchmarks show that compared with the baseline schemes, the average service latency of our scheme is reduced by more than 18.3%, and it has a higher convergence speed and task success rate, which can make ESs achieve load balancing.

With the rapid development of the Internet of Things (IoT), a massive amount of transient data is transmitted in edge networks.Transient data is highly time-sensitive, such as monitoring data generated by industrial devices.Due to their inefficiency, traditional caching strategies in edge networks are inadequate for handling transient data.Thus, to improve the efficiency of transient data caching, we construct a freshness model of transient data and propose a maximum entropy Actor-Critic based caching strategy, TD-MEAC-which can improve the freshness of cached data and reduce the long-term caching cost.Simulation results show that the proposed TD-MEAC achieves a higher cache hit rate and maintains a higher average freshness of cached transient data compared with the existing DRL and baseline caching strategies.

Log parsing, which converts semi-structured logs into structured logs, is the first step for automated log analysis. Existing parsers are still unsatisfactory in real-world systems due to new log types in new-coming logs. In practice, available logs collected during system runtime often do not contain all the possible log types of a system because log types related to infrequently activated system states are unlikely to be recorded and new log types are frequently introduced with system updates. Meanwhile, most existing parsers require preprocessing to extract variables in advance, but preprocessing is based on the operator’s prior knowledge of available logs and therefore may not work well on new log types. In addition, parser parameters set based on available logs are difficult to generalize to new log types. To support new log types, we propose a variable generation imitation strategy to craft a novel log parsing approach with generalization ability, called Log3T. Log3T employs a pre-trained transformer encoder-based model to extract log templates and can update parameters at parsing time to adapt to new log types by a modified test-time training. Experimental results on 16 benchmark datasets show that Log3T outperforms the state-of-the-art parsers in terms of parsing accuracy. In addition, Log3T can automatically adapt to new log types in new-coming logs.

In mobile edge computing environment, intelligent inference services driven by DNN are highly sensitive to latency. Recently, collaborative inference between User Devices and Edge Servers (ESs) based on Deep Neural Networks (DNN) partition has achieved success in service acceleration. However, most of the existing collaborative acceleration schemes are partitioned for a single DNN inference task, which cannot quickly make partition decisions for a set of concurrent inference tasks, and often sacrifice inference accuracy. In addition, due to the limited resources of ESs, there is resource competition among concurrent requests, which makes the partitioned tasks cannot be offloaded to ESs in time for processing. Therefore, designing an efficient offloading scheme becomes essential. The task offloading schemes based on deep reinforcement learning can solve complex decision-making problems in high-dimensional state space, but they have problems such as insufficient sample diversity and easily falling into local optimum. In this paper, a Collaborative Inference Acceleration Scheme integrating DNN Partitioning and Task Offloading (CIAS-PnO) is proposed. First, while ensuring inference accuracy, the Collaborative DNN Layer Partitioning (CDLP) algorithm is designed with the goal of optimal latency. CDLP can reduce the problem scale of concurrent inference tasks partition by pruning operation and determine the partition decisions in time. Then, the Distributed Soft Actor-Critic (SAC)-based Partition Task Offloading algorithm (DSACO) is designed. DSACO supports SAC Agents to explore samples in parallel and share learning experiences, and uses the automatic entropy adjustment mechanism to improve the exploration efficiency of Agents, so as to avoid falling into local optimum and achieve efficient offloading of partition tasks. Experimental results on DNN benchmarks show that compared with the baseline acceleration schemes, CIAS-PnO achieves more than 19.8% acceleration performance improvement, and has higher convergence performance and task success rate.

Automated log analysis can facilitate failure diagnosis for developers and operators using a large volume of logs. Log parsing is a prerequisite step for automated log analysis, which parses semi-structured logs into structured logs. However, existing parsers are difficult to apply to software-intensive systems, due to their unstable parsing accuracy on various software. Although neural network-based approaches are stable, their inefficiency makes it challenging to keep up with the speed of log production. In this work, we found that the longest common pattern among logs is likely to be part of the log template. Inspired by this key insight, we propose a new stable log parsing approach, called Brain, which creates initial groups according to the longest common pattern. Then a bidirectional tree is used to hierarchically complement the constant words to the longest common pattern to form the complete log template efficiently. Experimental results on 16 benchmark datasets show that our approach outperforms the state-of-the-art parsers on two widely-used parsing accuracy metrics, and it only takes around 46 seconds to process one million lines of logs.

To meet the low delay requirements for data content access in Industrial Internet of Things, efficient content caching strategies need to be designed in mobile edge computing architectures. Most existing caching strategies reduce access delay by predicting content popularity and caching popular content earlier during off-peak traffic, but these strategies only focus on the temporal features of content popularity and ignore the spatial correlation of content popularity. Therefore, we propose a mobile edge cooperative caching strategy based on spatio-temporal graph convolutional model(STCC). In STCC, we integrate graph convolutional neural network and the gated recurrent unit to mine the spatio-temporal characteristics of content popularity and make effective prediction. Moreover, we divide the collaboration domain by hierarchical clustering and design a heuristic cache placement strategy to minimize the access delay. Simulation experiments show that the spatio-temporal graph convolution model proposed in STCC can predict content popularity better than existing time series model. Compared with existing caching strategies, STCC can significantly improve the cache hit ratio and reduce the average content access delay.

Serverless computing/Function as a Service (FaaS) has emerged as a new paradigm for running short-lived applications in the cloud. Serverless edge computing is recently adopting serverless computing at edge to run event-driven tasks and supporting application running of resource-constrained Internet of Things (IoT) devices by offloading their tasks to the edge. However, traditional task offloading methods are mainly based on heuristic algorithms for one-shot optimization, which leads to performance degradation in long-term operation. Fortunately, deep reinforcement learning techniques combining reinforcement learning and deep neural networks provide a promising alternative. Therefore, a function offloading algorithm DRLFO is proposed with a deep reinforcement learning algorithm based on actor-critic framework in this paper. The function offloading process in serverless edge computing environment is modeled as a Markov Decision Process. Finally, the experimental results show that the proposed algorithm can successfully converge and outperform the compared baseline algorithm in terms of function success rate and reduce the average latency by 4.6%-22.6%.

Defect detection aims to locate the accurate position of defects in images, which is of great significance to quality inspection in the industrial product manufacturing. Currently, many defect detection methods rely on deep neural networks to extract features. Although the accuracy of these methods is relatively high, it is computationally intensive, making the methods difficult to deploy in resource-limited edge devices. In order to solve these problems, a lightweight defect detection model for the industrial edge environment is proposed, termed the efficient defect detection network (EDDNet). EfficientNet-B0 is used as the feature extraction backbone, extracting feature maps from feature layers of different depths of the network and fusing multilevel features by multilevel feature fusion (MFF). To obtain more information, we redesign the attention mechanism in MBConv blocks, taking the encoding space (ES) attention mechanism as a new module, which solves the problem that the defective image spatial information is ignored. The experimental results on the NEU-DET and DAGM2007 datasets and PCB defect datasets demonstrate the effectiveness of the proposed EDDNet and its possibility for application in industrial edge device.

This paper introduces a new Byzantine fault tolerance protocol called workload-based randomization Byzantine fault tolerance protocol (WRBFT). Improvements are made to the Practical Byzantine Fault Tolerance (PBFT), which has an important position in the Byzantine Fault consensus algorithm. Although PBFT has numerous advantages, its primary node selection mechanism is overly fixed, the communication overhead of the consensus process is also high, and nodes cannot join and exit dynamically. To solve these problems, the WRBFT proposed in this paper combines node consensus workload and verifiable random function (VRF) to randomly select the more reliable primary node that dominates the consensus. The selection of the nodes involved in the consensus is based on the node workload, and the optimization of the agreement protocol of the PBFT is also based on this. Simulation results show that the WRBFT has higher throughput, lower consensus latency, and higher algorithmic efficiency compared to the PBFT.

Addressing the problem of imbalanced data category distribution in real applications and the problem of traditional classifiers tending to ensure the accuracy of the majority class while ignoring the accuracy of the minority class when processing imbalanced data, this paper proposes a method called RBSP-Boosting for imbalanced data classification. First, RBSP-Boosting introduces the Shapley value and calculates the Shapley value for each sample of the dataset through the truncated Monte Carlo method. Moreover, the proposed method removes the noise data according to the Shapley value and undersamples the samples with Shapley values less than zero in the majority class. Then, it takes the Shapley value as the weight of the sample and oversamples the minority class according to the weight. Finally, the new dataset is trained on the classifier through the AdaBoost classifier. Experiments are conducted on nine groups of UCI and KEEL datasets, and RBSP-Boosting is compared with four sampling algorithms: Random-OverSampler, SMOTE, Borderline-SMOTE and SVM-SMOTE. Experimental results show that the RBSP-Boosting method in the three evaluation metrics of AUC, F-score and G-mean, compared with the best performance of the four comparison algorithms, increases by 4.69%, 10.3% and 7.86%, respectively. The proposed method can significantly improve the effect of imbalanced data classification.

Containers are light, numerous, and interdependent, which are prone to cascading fault, increasing the probability of fault and the difficulty of detection. Fault detection methods used in traditional cloud platforms will result in degraded detection performance when applied to the containerized cloud platform. This paper argues that there are usually two reasons that constrain the effectiveness of current cloud fault detection: (1) fault type level: the complexity of the cloud leads to a diversity of fault, many fault types are difficult to detect, such as container cascading fault; (2) fault data level: the imbalance of fault number data to the detection method to bring interference. Therefore, this paper proposes a Container cascading fault Detection strategy based on Spatial-temporal correlation model and Collaborative optimization(CDSC), by extracting the temporal and spatial correlation semantics between faulty containers, constructing the container cascade correlation model, and using dynamic feedback sampling combined with multi-model cooperative optimization training to improve the detection effect of container cascading fault. It is proved that this method improves the detection precision, recall rate and F1 value by 10%-15% on average compared with existing methods.

This paper first analyzes the Spread scheduling algorithm of the Docker orchestration engine SwarmKit. For the Spread algorithm to ignore the resource heterogeneity of the node, the application performance is not ideal when the resources are inconsistent. The RPBG strategy with collaborative orchestration is proposed. It mainly combines the steps of collaboratively initializing resource placement, dynamically evaluating dominant resources of service, and automatically adjusting node load to comprehensively consider the heterogeneity of resources such as memory, CPU and network I/O of each node of the cluster and resolve container conflicts based on graph theory. Finally, a series of experiments have been carried out to verify that RPBG is superior to Spread strategy in terms of resource utilization and application performance. The experimental results show that the application performance of the SwarmKit cluster with built-in RPBG strategy is improved by nearly 13% compared to the original strategy.

MOTIVATION: K-mers along with their frequency have served as an elementary building block for error correction, repeat detection, multiple sequence alignment, genome assembly, etc., attracting intensive studies in k-mer counting. However, the output of k-mer counters itself is large; very often, it is too large to fit into main memory, leading to highly narrowed usability. RESULTS: We introduce a novel idea of encoding k-mers as well as their frequency, achieving good memory saving and retrieval efficiency. Specifically, we propose a Bloom filter-like data structure to encode counted k-mers by coupled-bit arrays-one for k-mer representation and the other for frequency encoding. Experiments on five real datasets show that the average memory-saving ratio on all 31-mers is as high as 13.81 as compared with raw input, with 7 hash functions. At the same time, the retrieval time complexity is well controlled (effectively constant), and the false-positive rate is decreased by two orders of magnitude. AVAILABILITY AND IMPLEMENTATION: The source codes of our algorithm are available at github.com/lzhLab/kmcEx. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

The resource scheduling problem of data center clusters has always been a hot topic in the field of cloud computing. Existing research efforts focus on fairness, resource utilization and energy efficiency, and lack of research on heterogeneous clustering issues. To solve the problem that the traditional DRF algorithm does not consider the classification of machine performance and task type, this paper proposes a fair scheduling algorithm X-DRF that adapts to heterogeneous resources in the data center. The algorithm mainly classifies the performance of physical machines, increases the machine performance scoring factor, and increases the training and job type judgment classification of the XGBoost model. The experiments show that CPU utilization and memory usage increased by 10% and 6%, respectively. The normalized ratio is increased by about 3% compared to the original DRF system. Therefore, the presented fair scheduling algorithm for heterogeneous resources is more fair and reasonable in terms of resource allocation.

PaaS is one of the most popular paradigms of cloud computing and the performance guarantee of PaaS-oriented applications has been critically concerned. Performance models, such as a Layer Queue Network (LQN) model, are efficient at performance guaranteeing in a highly dynamic computing environment for their capabilities of capacity planning. However, it is not a trivial work to build and employ such models for a PaaS platform because of the lack of designs and the transaction-intensive feature of the PaaS-oriented applications. In this paper, a PaaS-oriented performance modeling approach is proposed. The LQN model of the PaaS-oriented application can be dynamically built through tracing their interactions with the PaaS platform. And the CPU consumptions, which are important to the accuracy of the LQN model, can be refined through a Kalman-filter method. A modeling tool is implemented and experimental results have shown the effectiveness of our approach.

Recognizing the human activities of daily living (ADL) is an important research issue in the pervasive environment. Activity recognition is treated as a classification problem and the multi-class classifier is often used. Though the multi-class classifier can obtain high classification accuracy, it can not detect the noise activities and unknown activities, and the system has no extendable recognition capability. In this paper, we proposed a recognition system which can recognize known activities and detect unknown activities simultaneously. For each known activity, one one-class classification model is built up and the combined one-class classification models are used to judge whether a test sample belongs to known activities. For the known samples, the multi-class classifier is used to recognize their types. For the continuous unknown samples, based on segmentation algorithm, training samples of new activities are extracted and added into the recognition system to extend the system's recognition capability.

Web services, with an emphasis on open standards and flexibility, can provide benefits over existing capital markets integration practices. However, Web services must first meet certain technical requirements including performance, security and so on. SOAP, based on Extensible Markup Language (XML), inherits not only the advantages of XML, but its relatively poor performance. This makes SOAP a poor choice for many high-performance Web services. In this paper, we propose a new approach to improve Web services performance. Focusing on avoiding traditional XML parsing and Java reflection at runtime, we create a service-specific SOAP processor to accelerate execution. Through our experiments in this paper, we observed that our approach obtained about a treble performance gain (maximum) by incorporating the SOAP processor into the SOAP engine

Web services are considered as solution for solving the interoperability problem and the challenge of integration. How to manage Web services more efficiently is a key problem to the applications based on Web services at present. This paper presents a framework for application server based Web services management named FASWSM. The main contribution of FASWSM is Web service adaptation, so that a Web service can be plugged into the application server (AS) as a Web service adapter which enables a Web service to be managed by the application server. Therefore, the Web services management efforts are greatly reduced by leveraging various services provided by the AS. The FASWSM also provides mechanisms for dynamical reconfiguration which improve the extensibility and flexibility of Web services management. The FASWSM has been applied to OnceAS 2.0 application server, and the experiments data show that FASWSM outperforms JAX-RPC in terms of Web service invocation.

As a basic reliability guarantee technology in distributed systems, failure detection provides the ability of timely detecting the liveliness of runtime systems. Effective failure detection is very important to J2EE application server (JAS), the leading middleware in Web computing environment, and it also needs to meet the requirements of reconfiguration, flexibility and adaptability. Based on the QoS (quality of service) specification of failure detector, this paper presents a QoS-enable failure detection framework for JAS, which satisfies the requirements of dynamically adjusting qualities and flexible integration of failure detectors. The work has been implemented in OnceAS application server that is developed by Institute of Software, Chinese Academy of Sciences. The experiments show that the framework can provide good QoS of failure detection in JAS.