The successful deployment of Digital Twins for real-time optimisation of physical systems relies critically on highly accurate and efficient deep learning surrogate models. Ensuring these models meet performance and latency requirements demands rigorous Neural Architecture Search (NAS) and Hyperparameter Optimisation (HPO). While both have been traditionally posed as a pure-exploration bandit problem, we show that it fails to capture the unique, deterministic characteristics of Scientific Machine Learning (SciML) models underpinning Digital Twins, particularly Physics-Informed Neural Networks (PINNs) and Deep Operator Networks (DeepONets). We propose a non-stochastic multi-armed bandit with balanced exploration-exploitation as the proper setting for NAS in SciML and introduce BanditNAS , a novel algorithm that addresses three critical challenges absent from current approaches: (i) late convergence in high-capacity models exhibiting spectral bias, (ii) validation loss plateaus requiring optimiser switching and (iii) the deterministic (non-stochastic) nature of physics-based training data. We analyse BanditNAS ’s theoretical properties, proving improved regret bounds compared to adaptive adversaries and compare it empirically with state-of-the-art approaches across three representative SciML scenarios. Our results demonstrate setting-dependent performance: BanditNAS achieves up to \(95\%\) higher optimal selection rates when multi-stage fine-tuning is required (DeepONets with L-BFGS switching), approximately \(50\%\) improvement in late-convergence regimes (high-capacity PINNs) and comparable performance to HyperBand in moderately noisy environments, although underperforming in very large search spaces with high noise ( \(K=200\) graph networks). Statistical significance testing confirms BanditNAS ’s superiority in two of three settings ( \(p < 0.001\) ), with competitive performance in the third when restricted to \(K\leq 100\) . These findings establish BanditNAS as a viable and theoretically grounded approach for optimising SciML models where real-time accuracy, multi-stage training and computational resource constraints are paramount, while highlighting the importance of algorithm selection based on problem characteristics.

Abstract Siamese Neural Networks (SNNs) have shown promise in addressing a variety of tasks, even with limited data availability. However, their adoption is hindered by the lack of transparency in their decision-making processes. A key challenge in explaining SNNs lies in the absence of an inverse mapping between high-dimensional input feature vectors and the low-dimensional embedding space. Therefore, computing direct distances between input features becomes meaningless. Existing autoencoder-based explanation methods face several limitations. These include poor image reconstruction quality due to insufficient data and the omission of final distance layer of the SNN during the explanation process. While the Siamese Network Explainer (SINEX) can explain audio and grayscale images, it does not support RGB images. To overcome these challenges, we propose a method called Features Distance-based eXplanation (FDbX). This approach identifies salient features using ridge regression, trained on perturbed SLIC-segmented images. To enhance the selection of important features, we incorporate Bayesian analysis, which assigns importance scores to features. To provide a comprehensive explanation of the decision route, we construct a mathematical model that represents important features and their Hamming distances as a bipartite graph. In this graph, nodes represent features and edges denote distances between feature pairs. The resulting explanation heatmaps highlight critical image segments, offering more intuitive and visually informative explanations than existing methods. We evaluate stability and faithfulness of our method using stability indices such as $$R^2$$ and mean squared error. To the best of our knowledge, this is the first work to introduce Variable and Coefficient Stability Indices for image datasets.

Collaborative blockchain storage is one of the solutions to the blockchain storage scalability challenge, allowing individual nodes to store partial blockchain data. Collaborative blockchains give up redundancy for scalability. However, redundancy is a determinant of blockchain’s decentralization and data availability. In this article, we study the possibility of balancing the scalability–redundancy tradeoff in collaborative blockchains. We first present a model for the evaluation of a data distribution scheme for its efficiency in terms of redundancy. Then, we present a fully decentralized self-regulated data distribution scheme assuring storage scalability and redundancy for collaborative blockchains based on fully redundant global state. The proposed solution does not incur additional communication. We examine the efficiency of the proposed scheme based on experiments. Our results demonstrate that the proposed system assures a level of redundancy for a given set of nodes and data volume, optimizes storage space utilization, and provides uniform distribution of data.

Abstract Physics Informed Machine Learning has emerged as a popular approach for modeling and simulation in digital twins, enabling the generation of accurate models of processes and behaviors in real-world systems. However, existing methods either rely on simple loss regularizations that offer limited physics integration or employ highly specialized architectures that are difficult to generalize across diverse physical systems. This paper presents a generic approach based on a novel physics-encoded residual neural network (PERNN) architecture that seamlessly combines data-driven and physics-based analytical models to overcome these limitations. Our method integrates differentiable physics blocks–implementing mathematical operators from physics-based models–with feed-forward learning blocks, while intermediate residual blocks ensure stable gradient flow during training. Consequently, the model naturally adheres to the underlying physical principles even when prior physics knowledge is incomplete, thereby improving generalizability with low data requirements and reduced model complexity. We investigate our approach in two application domains. The first is a steering model for autonomous vehicles in a simulation environment, and the second is a digital twin for climate modeling using an ordinary differential equation (ODE)-based model of Net Ecosystem Exchange (NEE) to enable gap-filling in flux tower data. In both cases, our method outperforms conventional neural network approaches as well as state-of-the-art Physics Informed Machine Learning methods.

Malicious Software (Malware) is a primary element of many cyber crimes and attacks, causing massive damage and financial losses to organizations. Accordingly, malware detection and classification has become a crucial security field resulting in various attempts from researchers to develop solutions including signature-based approaches to Artificial Intelligence (AI) models showing their efficacy in detecting malware. Yet, users still have reservations about AI models due to the ambiguity and mysteriousness of their decisions resulting from their black-box nature. To address this problem, this paper proposes to develop explainable AI models to classify the malware families robustly. The proposed method uses text classification of API call sequences generated by these families by considering two datasets. A weighted training methodology is used to solve the dataset imbalance problem. Subsequently, the method presents an eXplainable AI (XAI) approach to establish an understandable and interpretable relationship between the API call sequences and the Bidirectional Encoder Representations from Transformers (BERT) model decisions, which enhance the model accountability and usability by employing the Local Interpretable Model-Agnostic Explanation (LIME) and the Shapley Additive Explanations (SHAP) platforms. The results reveal that the BERT model outperforms its counterparts considering F1 score, Balanced Accuracy (BA), and Matthews correlation coefficient (MCC).

SUMMARY The growing number of services processed and stored in the cloud has led to difficulties in managing and discovering the required services efficiently. Multilevel index model is an efficient method to manage and retrieve services in service repositories. When adding a new service to a multilevel index model, a key needs to be selected for the service, but existing key selection methods cannot adapt to the situation that hot services change over time. To address this problem, this article proposes an adaptive key selection method to improve the efficiency of service retrieval. However, the service addition operation of the adaptive key selection method is inefficient in the multilevel index model. For this reason, this article improves the multilevel index model by introducing local equivalence partition. This indexing model improves the service addition efficiency of the adaptive key selection method without affecting the service retrieval efficiency. It is experimentally demonstrated that the retrieval and addition efficiencies of the adaptive key selection method are close to the ideal state optimum under the multilevel index model with local equivalence partitioning.

The distributed deep learning architecture can support the front-deployment of deep learning systems in resource constrained IoT devices and is attracting increasing interest. However, most ready-to-use deep models are designed for centralized deployment without considering the transmission loss of the intermediate representation inside the distributed architecture. This oversight significantly affects the inference performance of distributed deployed deep models. To alleviate this problem, a state-of-the-art work chooses to retrain the original model to form an intermediate representation with ordered importance and yields better inference accuracy under constrained transmission bandwidth. This paper first reveals that this solution is essentially a pruning-like solution, where unimportant information is adaptively pruned to fit within the limited bandwidth. With this understanding, a novel scheme named Naturally Aggregated Intermediate Representation (NAIR) has been proposed, which aims to naturally amplify the difference of importance embedded in the intermediate representation from a mature deep model and reassemble the intermediate representation into a hierarchy of importance from high-to-low to accommodate the transmission loss. As a result, this method shows further improved performance in various scenarios, avoids compromising the overall inference performance of the system, and saves astronomical retraining and storage costs. The effectiveness of NAIR has been validated through extensive experiments, achieving a 112% improvement in performance compared to the state-of-the-art work.

Building upon pre-trained ViT models, many advanced methods have achieved significant success in COVID-19 classification. Many scholars pursue better performance by increasing model complexity and parameters. While these methods can enhance performance, they also require extensive computational resources and extended training times. Additionally, the persistent challenge of overfitting, due to limited COVID-19 dataset sizes, remains a hurdle. To address these challenges, we proposed a novel method to optimize pre-trained transformer models for efficient COVID-19 classification with stochastic configuration networks (SCNs), referred to as OPT-CO. We proposed two optimization methods: sequential optimization (SeOp) and parallel optimization (PaOp), by incorporating optimizers in a sequential and parallel manner, respectively. Our method can enhance model performance without necessitating a significant parameter expansion. Additionally, we introduced OPT-CO-SCN to avoid overfitting problems through the adoption of random projection for head augmentation. The experiments were carried out to evaluate the performance of our proposed model based on two publicly available datasets. Based on the evaluation results, our method achieved superior, performance surpassing other state-of-the-art methods.

Abstract With the development of edge computing and artificial intelligence (AI) technologies, edge devices are witnessed to generate data at unprecedented volume. The edge intelligence (EI) has led to the emergence of edge devices in various application domains. The EI can provide efficient services to delay‐sensitive applications, where the edge devices are deployed as edge nodes to host the majority of execution, which can effectively manage services and improve service discovery efficiency. The multilevel index model is a well‐known model used for indexing service, such a model is being introduced and optimized in the edge environments to efficiently services discovery while managing large volumes of data. However, effectively updating the multilevel index model by adding new services timely and precisely in the dynamic edge computing environments is still a challenge. Addressing this issue, this article proposes a designated key selection method to improve the efficiency of adding services in the multilevel index models. Our experimental results show that in the partial index and the full index of multilevel index model, our method reduces the service addition time by around 84% and 76%, respectively when compared with the original key selection method and by around 78% and 66%, respectively when compared with the random selection method. Our proposed method significantly improves the service addition efficiency in the multilevel index model, when compared with existing state‐of‐the‐art key selection methods, without compromising the service retrieval stability to any notable level.

Abstract The processing of locally harvested data at the physically accessible edge devices opens a new avenue of security threats for edge enhanced analytics. Cryptographic algorithms are used to secure the data being processed on the edge device. However, the implementation weakness of the algorithms on the edge devices can lead to side‐channel attack vulnerability, which is exacerbated with the application of machine‐learning techniques. This research proposes a deep learning‐based system integrated at the edge device to identify the side‐channel leakages. To design such a deep learning‐based system, one of the challenges is formulating the suitable attack model for the underlying target algorithm. Based on the previous findings, three machine learning‐based side‐channel attack models are curated and investigated for the edge device security evaluations. As a test case, the standard elliptic‐curve cryptographic algorithm is selected. Moreover, quantitative analysis is provided for the best attack model selection using standard machine‐learning evaluation metrics. A comparative analysis is performed on the raw unaligned data samples and reduced feature‐engineered samples using edge enhanced security analytics. The investigation concludes that the vulnerable algorithm implementation can lead to the secret key recovery from the edge device, with 96% accuracy, using a neural‐network‐based algorithm to analyse side‐channel attacks.

With the advent of the Internet of everything era, people's demand for intelligent Internet of Things (IoT) devices is steadily increasing. A more intelligent cyber-physical system (CPS) is needed to meet the diverse business requirements of users, such as ultra-reliable low-latency communication, high quality of services (QoS), and quality of experience (QoE). Edge intelligence (EI) is recognized by academia and industry as one of the key emerging technologies for the CPS, which provides the ability to analyze data at the edge rather than sending it to the cloud for analysis, and will be a key enabler to realize a world of a trillion hyperconnected smart sensing devices.As a distributed intelligent computing paradigm in which computation is largely or completely performed at distributed nodes, EI provides for the rapid development of artificial intelligence (AI) and edge computing resources to support real-time insight and analysis for applications in CPS, which brings memory, computing power and processing ability closer to the location where it is needed, reduces the volumes of data that must be moved, the consequent traffic, and the distance the data must travel. As an emerging intelligent computing paradigm, EI can accelerate content delivery and improve the QoS of applications, which is attracting more and more research attentions from academia and industry because of its advantages in throughput, delay, network scalability and intelligence in CPS.

Existing Cloud systems involve large volumes of data streams being sent to a centralised data centre for monitoring, storage and analytics. However, migrating all the data to the cloud is often not feasible due to cost, privacy, and performance concerns. However, Machine Learning (ML) algorithms typically require significant computational resources, hence cannot be directly deployed on resource-constrained edge devices for learning and analytics. Edge-enhanced compressive offloading becomes a sustainable solution that allows data to be compressed at the edge and offloaded to the cloud for further analysis, reducing bandwidth consumption and communication latency. The design and implementation of a learning method for discovering compression techniques that offer the best QoS for an application is described. The approach uses a novel modularisation approach that maps features to models and classifies them for a range of Quality of Service (QoS) features. An automated QoS-aware orchestrator has been designed to select the best autoencoder model in real-time for compressive offloading in edge-enhanced clouds based on changing QoS requirements. The orchestrator has been designed to have diagnostic capabilities to search appropriate parameters that give the best compression. A key novelty of this work is harnessing the capabilities of autoencoders for edge-enhanced compressive offloading based on portable encodings, latent space splitting and fine-tuning network weights. Considering how the combination of features lead to different QoS models, the system is capable of processing a large number of user requests in a given time. The proposed hyperparameter search strategy (over the neural architectural space) reduces the computational cost of search through the entire space by up to 89%. When deployed on an edge-enhanced cloud using an Azure IoT testbed, the approach saves up to 70% data transfer costs and takes 32% less time for job completion. It eliminates the additional computational cost of decompression, thereby reducing the processing cost by up to 30%.

A substantial amount of research has been carried out in Explainable Artificial Intelligence (XAI) models, especially in those which explain the deep architectures of neural networks. A number of XAI approaches have been proposed to achieve trust in Artificial Intelligence (AI) models as well as provide explainability of specific decisions made within these models. Among these approaches, global interpretation methods have emerged as the prominent methods of explainability because they have the strength to explain every feature and the structure of the model. This survey attempts to provide a comprehensive review of global interpretation methods that completely explain the behaviour of the AI models. We present a taxonomy of the available global interpretations models and systematically highlight the critical features and algorithms that differentiate them from local as well as hybrid models of explainability. Through examples and case studies from the literature, we evaluate the strengths and weaknesses of the global interpretation models and assess challenges when these methods are put into practice. We conclude the paper by providing the future directions of research in how the existing challenges in global interpretation methods could be addressed and what values and opportunities could be realized by the resolution of these challenges.

With increasing availability and use of Internet of Things (IoT) devices such as sensors and video cameras, large amounts of streaming data is now being produced at high velocity. Applications which require low latency response such as video surveillance, augmented reality and autonomous vehicles demand a swift and efficient analysis of this data. Existing approaches employ cloud infrastructure to store and perform machine learning-based analytics on this data. This centralized approach has limited ability to support real-time analysis of large-scale streaming data due to network bandwidth and latency constraints between data source and cloud. We propose RealEdgeStream (RES) an edge enhanced stream analytics system for large-scale, high performance data analytics. The proposed approach investigates the problem of video stream analytics by proposing (i) filtration and (ii) identification phases. The filtration phase reduces the amount of data by filtering low-value stream objects using configurable rules. The identification phase uses deep learning inference to perform analytics on the streams of interest. The phases consist of stages which are mapped onto available in-transit and cloud resources using a placement algorithm to satisfy the Quality of Service (QoS) constraints identified by a user. We demonstrate that for a 10K element data streams, with a frame rate of 15–100 per second, the job completion in the proposed system takes 49 percent less time and saves 99 percent bandwidth compared to a centralized cloud-only based approach.

Abstract: Recent advances in VR/AR have fueled the interest in automating the development of 3D interfaces. An important step of this automation is 3D object recognition. This allows applications to pre-configure assets based on expected behaviours, rather than requiring human developers to hard code such assets. This paper presents a clustering approach for reliably identifying 3D objects. Model-based descriptors are used to extract key features and geometric data are used to develop models for 3D object recognition. Two clustering approaches are used in this paper: hierarchical and K-means. The models are tested on two industrial gearboxes and an accuracy score is calculated to test model performance. The two gearboxes were too small resulting in the accuracy of both algorithms being identical, however the computational efficiency of the K-means algorithm makes it a more suitable choice. The accuracy for both algorithms was 74% when tested on a combined dataset of both gearboxes. This increased to 81% when difficult to identify parts are removed. The model can reliably identify geometrically consistent parts like ball bearings, roller bearings and screws.

With the Internet of Things technology, almost any remote sensing devices, wearables, and smart objects are equipped to transmit large volumes of data in continuous streams. In conventional cloud-centric analytics, all the raw data is transferred to a data centre and processed in real-time, near real-time, or in batches. However, this approach is usually not very responsive to real-time analytics due to the latency in transmission alongside network traffic, bandwidth and data transmission costs. To tackle this, edge-enhanced analytics ensures that raw data can be preprocessed at the edge and sent across the network channel in a more compact form. A specific category of deep learning model, autoencoder, can help to achieve this by transforming high-dimensional data into compact representation. We propose an edge-enhanced framework which deploys a deep autoencoder model on the network edge for data compression. After training of models in the cloud, the encoder part of the autoencoder is deployed on the edge for data reduction while the decoder remains on the cloud to reconstruct the data for an image classification task on the cloud. We applied supervised fine-tuning using the intrinsic dimensionality of the data to achieve an accuracy that surpasses the baseline cloud model. The solution was explored in the context of an image recognition problem using the MNIST and FASHION-MNIST datasets. The framework was validated on an event simulator to estimate the network savings of the proposed method in terms of bandwidth and latency. The edge-enhanced approach saves up to 74% bandwidth compared to the centralised analytics. In addition, real-time analytics is further improved by taking 25% less time to complete the task.

The diversity and proliferation of Knowledge bases have made data integration one of the key challenges in the data science domain. The imperfect representations of entities, particularly in graphs, add additional challenges in data integration. Graph dependencies (GDs) were investigated in existing studies for the integration and maintenance of data quality on graphs. However, the majority of graphs contain plenty of duplicates with high diversity. Consequently, the existence of dependencies over these graphs becomes highly uncertain. In this paper, we proposed graph probabilistic dependencies (GPDs) to address the issue of uncertainty over these large-scale graphs with a novel class of dependencies for graphs. GPDs can provide a probabilistic explanation for dealing with uncertainty while discovering dependencies over graphs. Furthermore, a case study is provided to verify the correctness of the data integration process based on GPDs. Preliminary results demonstrated the effectiveness of GPDs in terms of reducing redundancies and inconsistencies over the benchmark datasets.

Due to the popularity of onboard geographic devices, a large number of spatial-textual objects are generated in the Internet of Vehicles (IoV). This development calls for approximate spatial keyword queries with numeric attributes in IoV (A2SKIV), which takes into account the locations, textual descriptions, and numeric attributes of spatial-textual objects. Considering large amounts of objects involved in the query processing, this article comes up with the idea of utilizing vehicles as fog-computing resource and proposes the network structure called FCV, and based on which the fog-based top-k A2SKIV query is explored and formulated. In order to effectively support network distance pruning, textual semantic pruning, and numerical attribute pruning, simultaneously, a two-level spatial-textual hybrid index STAG-tree is designed. Based on STAG-tree, an efficient top-k A2SKIV query processing algorithm is presented. The simulation results show that our STAG-based approach is about 1.87× (17.1×, resp.) faster in search time than the compared ILM (DBM, resp.) method, and our approach is scalable.

Combining edge processing (at data capture site) with analysis carried out while data is enroute from the capture site to a data center offers a variety of different processing models. Such in-transit nodes include network data centers that have generally been used to support content distribution (providing support for data multicast and caching), but have recently started to offer user-defined programmability, through Software Defined Networks (SDN) capability, e.g., OpenFlow and Network Function Visualization (NFV). We demonstrate how this multi-site computational capability can be aggregated to support video analytics, with Quality of Service and cost constraints (e.g., latency-bound analysis). The use of SDN technology enables separation of the data path from the control path, enabling in-network processing capabilities to be supported as data is migrated across the network. We propose to leverage SDN capability to gain control over the data transport service with the purpose of dynamically establishing data routes such that we can opportunistically exploit the latent computational capabilities located along the network path. Using a number of scenarios, we demonstrate the benefits and limitations of this approach for video analysis, comparing this with the baseline scenario of undertaking all such analysis at a data center located at the core of the infrastructure.

Deep Reinforcement Learning (DRL) has been proven to be capable of designing an optimal control theory by minimising the error in dynamic systems. However, in many of the real-world operations, the exact behaviour of the environment is unknown. In such environments, random changes cause the system to reach different states for the same action. Hence, application of DRL for unpredictable environments is difficult as the states of the world cannot be known for non-stationary transition and reward functions.

As the streaming data generated by Internet of Things (IoT) ubiquitous sensors grow in massive scale, extracting interesting information (anomalies) in real-time becomes more challenging. Traditional systems which retrospectively perform all the processing in the cloud do not capture real-time changes in the data. Similarly, real-time solutions which rely on human monitors have the tendency to miss the anomalies due to their rare nature. In recent times, several machine learning techniques have been proposed for stream processing. Approaches based on supervised or semi-supervised learning fail to adapt to changing patterns of the streaming data and the data labelling costs are huge. To address these limitations, we propose a cloud-assisted framework where an intermediary node (edge) is introduced between the end devices and the cloud to assist in stream processing. A model deployed on the edge is designed to learn in an iterative manner to discriminate between similar and dissimilar data representations, making it easier to distinguish the anomalies. In this work, we have proposed an iterative method that combines the capabilities of deep clustering and l2-normalisation to achieve better discriminative representations. Experimental results demonstrate the proposed method achieves robust performance over state-of-the-art discriminative representation algorithms and sets new benchmark accuracy on transformation invariant image dataset.

Summary Object classification is a vital part of any video analytics system, which could aid in complex applications such as object monitoring and management. Traditional video analytics systems work on shallow networks and are unable to harness the power of distributed processing for training and inference. We propose a cloud‐based video analytics system based on an optimally tuned convolutional neural network to classify objects from video streams. The tuning of convolutional neural network is empowered by in‐memory distributed computing. The object classification is performed by comparing the target object with the prestored trained patterns, generating a set of matching scores. The matching scores greater than an empirically determined threshold reveal the classification of the target object. The proposed system proved to be robust to classification errors with an accuracy and precision of 97% and 96%, respectively, and can be used as a general‐purpose video analytics system.

Object detection and classification are the basic tasks in video analytics and become the starting point for other complex applications. Traditional video analytics approaches are manual and time consuming. These are subjective due to the very involvement of human factor. We present a cloud based video analytics framework for scalable and robust analysis of video streams. The framework empowers an operator by automating the object detection and classification process from recorded video streams. An operator only specifies an analysis criteria and duration of video streams to analyse. The streams are then fetched from a cloud storage, decoded and analysed on the cloud. The framework executes compute intensive parts of the analysis to GPU powered servers in the cloud. Vehicle and face detection are presented as two case studies for evaluating the framework, with one month of data and a 15 node cloud. The framework reliably performed object detection and classification on the data, comprising of 21,600 video streams and 175 GB in size, in 6.52 hours. The GPU enabled deployment of the framework took 3 hours to perform analysis on the same number of video streams, thus making it at least twice as fast than the cloud deployment without GPUs.

A system to perform video analytics is proposed using a dynamically tuned convolutional network. Videos are fetched from cloud storage, preprocessed, and a model for supporting classification is developed on these video streams using cloud-based infrastructure. A key focus in this paper is on tuning hyper-parameters associated with the deep learning algorithm used to construct the model. We further propose an automatic video object classification pipeline to validate the system. The mathematical model used to support hyper-parameter tuning improves performance of the proposed pipeline, and outcomes of various parameters on system's performance is compared. Subsequently, the parameters that contribute toward the most optimal performance are selected for the video object classification pipeline. Our experiment-based validation reveals an accuracy and precision of 97% and 96%, respectively. The system proved to be scalable, robust, and customizable for a variety of different applications.

A new method is presented for improvement of the particle identification analysis process in a way which combines both the measured features, from detectors, and physics parameters. It is proposed that a graph representation can effectively express data in a format allowing for simpler interpretation and exploitation of all data available for analysis purposes. Nodes will represent entities and edges will represent the relation between them. Not only are graphs able to provide this useful structure and formal representation of knowledge but they can also be managed efficiently. Overall, this graphical representation will allow for the study of relationships between tracks, enable better pattern recognition and, as a result, improve the classification of particles.

Applying deep learning models to large-scale IoT data is a compute-intensive task and needs significant computational resources. Existing approaches transfer this big data from IoT devices to a central cloud where inference is performed using a machine learning model. However, the network connecting the data capture source and the cloud platform can become a bottleneck. We address this problem by distributing the deep learning pipeline across edge and cloudlet/fog resources. The basic processing stages and trained models are distributed towards the edge of the network and on in-transit and cloud resources. The proposed approach performs initial processing of the data close to the data source at edge and fog nodes, resulting in significant reduction in the data that is transferred and stored in the cloud. Results on an object recognition scenario show 71\% efficiency gain in the throughput of the system by employing a combination of edge, in-transit and cloud resources when compared to a cloud-only approach.

Inter-cloud is an approach that facilitates scalable resource provisioning across multiple cloud infrastructures. In this paper, we focus on the performance optimization of Infrastructure as a Service (IaaS) using the meta-scheduling paradigm to achieve an improved job scheduling across multiple clouds. We propose a novel inter-cloud job scheduling framework and implement policies to optimize performance of participating clouds. The framework, named as Inter-Cloud Meta-Scheduling (ICMS), is based on a novel message exchange mechanism to allow optimization of job scheduling metrics. The resulting system offers improved flexibility, robustness and decentralization. We implemented a toolkit named “Simulating the Inter-Cloud” (SimIC) to perform the design and implementation of different inter-cloud entities and policies in the ICMS framework. An experimental analysis is produced for job executions in inter-cloud and a performance is presented for a number of parameters such as job execution, makespan, and turnaround times. The results highlight that the overall performance of individual clouds for selected parameters and configuration is improved when these are brought together under the proposed ICMS framework.

Video analytics systems based on deep learning approaches are becoming the basis of many widespread applications including smart cities to aid people and traffic monitoring. These systems necessitate massive amounts of labeled data and training time to perform fine tuning of hyper-parameters for object classification. We propose a cloud based video analytics system built upon an optimally tuned deep learning model to classify objects from video streams. The tuning of the hyper-parameters including learning rate, momentum, activation function and optimization algorithm is optimized through a mathematical model for efficient analysis of video streams. The system is capable of enhancing its own training data by performing transformations including rotation, flip and skew on the input dataset making it more robust and self-adaptive. The use of in-memory distributed training mechanism rapidly incorporates large number of distinguishing features from the training dataset - enabling the system to perform object classification with least human assistance and external support. The validation of the system is performed by means of an object classification case-study using a dataset of 100GB in size comprising of 88,432 video frames on an 8 node cloud. The extensive experimentation reveals an accuracy and precision of 0.97 and 0.96 respectively after a training of 6.8 hours. The system is scalable, robust to classification errors and can be customized for any real-life situation.

Ever since the completion of the Human Genome Project in 2003, the human genome has been represented as a linear sequence of 3.2 billion base pairs and is referred to as the "Reference Genome". Since then it has become easier to sequence genomes of individuals due to rapid advancements in technology, which in turn has created a need to represent the new information using a different representation. Several attempts have been made to represent the genome sequence as a graph albeit for different purposes. Here we take a look at the Variant Calling Format (VCF) file which carries information about variations within genomes and is the primary format of choice for genome analysis tools. This short paper aims to motivate work in representing the VCF file as Directed Acyclic Graphs (DAGs) to run on a cloud in order to exploit the high performance capabilities provided by cloud computing.

The recent rise in multimedia technology has made it easier to perform a number of tasks. One of these tasks is monitoring where cheap cameras are producing large amount of video data. This video data is then processed for object classification to extract useful information. However, the video data obtained by these cheap cameras is often of low quality and results in blur video content. Moreover, various illumination effects caused by lightning conditions also degrade the video quality. These effects present severe challenges for object classification. We present a cloud-based blur and illumination invariant approach for object classification from images and video data. The bi-dimensional empirical mode decomposition (BEMD) has been adopted to decompose a video frame into intrinsic mode functions (IMFs). These IMFs further undergo to first order Reisz transform to generate monogenic video frames. The analysis of each IMF has been carried out by observing its local properties (amplitude, phase and orientation) generated from each monogenic video frame. We propose a stack based hierarchy of local pattern features generated from the amplitudes of each IMF which results in blur and illumination invariant object classification. The extensive experimentation on video streams as well as publically available image datasets reveals that our system achieves high accuracy from 0.97 to 0.91 for increasing Gaussian blur ranging from 0.5 to 5 and outperforms state of the art techniques under uncontrolled conditions. The system also proved to be scalable with high through-put when tested on a number of video streams using cloud infrastructure.

Summary The evolution of communication protocols, sensory hardware, mobile and pervasive devices, alongside social and cyber‐physical networks, has made the Internet of things (IoT) an interesting concept with inherent complexities as it is realised. Such complexities range from addressing mechanisms to information management and from communication protocols to presentation and interaction within the IoT. Although existing Internet and communication models can be extended to provide the basis for realising IoT, they may not be sufficiently capable to handle the new paradigms that IoT introduces, such as social communities, smart spaces, privacy and personalisation of devices and information, modelling and reasoning. With interaction models in IoT moving from the orthodox service consumption model, towards an interactive conversational model, nature‐inspired computational models appear to be candidate representations. Specifically, this research contests that the reactive and interactive nature of IoT makes chemical reaction‐inspired approaches particularly well suited to such requirements. This paper presents a chemical reaction‐inspired computational model using the concepts of graphs and reflection, which attempts to address the complexities associated with the visualisation, modelling, interaction, analysis and abstraction of information in the IoT. Copyright © 2013 John Wiley & Sons, Ltd.

Ant Colony Optimization (ACO) and other similar nature inspired mechanisms like artificial neural networks, swarm intelligence and evolutionary algorithms are based on naturally existing Complex Adaptive Systems (CAS). Human immune system, sand dune ripples, and ant foraging are some examples of the naturally existing CAS. Participating agents in these systems interact according to simple local rules which result in complex behavior and self-organization at system level. Adhoc grids are dynamic in nature and participating nodes show intermittent and volatile participation. Resource availability fluctuates over time inadhoc grids and results in a new adhoc grid state. These changes require adoption of the adhoc grid to anew state by applying some self organizing mechanism. In this paper, we present nature-inspired (ACO), micro-economic based mechanisms for infrastructure level self-organization in adhoc grids. These mechanisms help in achieving a scalable, dynamic and a self-organizing adhoc grid infrastructure. These mechanisms are evaluated with varying workloads in different network conditions. Study of these mechanisms helped in understanding the effect of ACO based self-organization mechanism on the infrastructural spectrum, ranging from completely centralized to fully decentralized.

'Simulating the Inter-Cloud' (SimIC) is a discrete event simulation toolkit based on the process oriented simulation package of SimJava. The SimIC aims of replicating an inter-cloud facility wherein multiple clouds collaborate with each other for distributing service requests with regards to the desired simulation setup. The package encompasses the fundamental entities of the inter-cloud meta-scheduling algorithm such as users, meta-brokers, local-brokers, datacenters, hosts, hyper visors and virtual machines (VMs). Additionally, resource discovery and scheduling policies together with VMs allocation, re-scheduling and VM migration strategies are included as well. Using the SimIC a modeler can design a fully dynamic inter-cloud setting wherein collaboration is founded on meta-scheduling inspired characteristics of distributed resource managers that exchange user requirements as driven events in real-time simulations. The SimIC aims of achieving interoperability, flexibility and service elasticity while at the same time introducing the notion of heterogeneity of multiple clouds' configurations. In addition it accepts an optimization of a variety of selected performance criteria for a diversity of entities. The crucial factor of dynamics consideration has implemented by allowing reactive orchestration based on current workload of already executed heterogeneous user specifications. These are in the form of text files that the modeler can load in the toolkit and occurs in real-time at different simulation intervals. Finally, a unique request is scheduled for execution to an internal cloud datacenter host VM that is capable of performing the service contract. This is formally designed in Service Level Agreements (SLAs) based upon user profiling.

Scientific workflows have become the primary mechanism for conducting analyses on distributed computing infrastructures such as grids and clouds. In recent years, the focus of optimization within scientific workflows has primarily been on computational tasks and workflow makespan. However, as workflow-based analysis becomes ever more data intensive, data optimization is becoming a prime concern. Moreover, scientific workflows can scale along several dimensions: (i) number of computational tasks, (ii) heterogeneity of computational resources, and the (iii) size and type (static versus streamed) of data involved. Adapting workflow structure in response to these scalability challenges remains an important research objective. Understanding how a workflow graph can be restructured in an automated manner (through task merge, for instance), to address constraints of a particular execution environment is explored in this work, using a multi-objective evolutionary approach. Our approach attempts to adapt the workflow structure to achieve both compute and data optimization. The question of when to terminate the evolutionary search in order to conserve computations is tackled with a novel termination criterion. The results presented in this article demonstrate the feasibility of the termination criterion and demonstrate that significant optimization can be achieved with a multi-objective approach.

The ability to seamlessly bridge clouds across organisational and administrative boundaries will play a vital role in establishing the utility of cloud computing for large-scale collaborative processes. Managing human and environmental contexts across geographical, network and administrative boundaries is a process that can benefit from a federation of cloud platforms. In the absence of a mature standard that defines access, control, management and coordination mechanisms between clouds in a federation, we explore these issues through a use-case of managing the dissemination and consumption of contextual information. The use-case is driven by the deployment of a broker-based context provisioning system, for homogenous cloud deployments that reside in different administrative domains. The discussion is driven by the aim to highlight key issues and challenges for enabling cloud federation for large-scale context provisioning, which forms the main contribution of this article.

Large-scale neuroscience research projects are necessary in order to make significant progress in the study of degenerative brain diseases. At present the effectiveness of such efforts is being somewhat restricted by the absence of specifically tailored computing infrastructures. The neuGRID project aims to address this through the provision of a high-level service oriented infrastructure that enables complex neuro-science research. One of the principle aims of this work is to develop portable services that can be re-used in a larger set of related medical applications to access distributed computing resources. These services will provide high-level functionality that will support workflow authoring and planning, provenance storage and retrieval, querying against heterogeneous data sources as well as security and data anonymization amongst others. This paper introduces the neuGRID service architecture and outlines the design of two specific services, namely the Pipeline Service and the Glueing Service. A proof of concept implementation to evaluate the neuGRID design approach has been developed.

Evidence-based medicine is critically dependent on three sources of information: a medical knowledge base, the patient's medical record and knowledge of available resources, including where appropriate, clinical protocols. Patient data is often scattered in a variety of databases and may, in a distributed model, be held across several disparate repositories. Consequently addressing the needs of an evidence- based medicine community presents issues of biomedical data integration, clinical interpretation and knowledge management. This paper outlines how the Health-e-Child project has approached the challenge of requirements specification for (bio-) medical data integration, from the level of cellular data, through disease to that of patient and population. The approach is illuminated through the requirements elicitation and analysis of Juvenile Idiopathic Arthritis (JIA), one of three diseases being studied in the EC-funded Health- e-Child project.

The use of meta-schedulers for resource management in large-scale distributed systems often leads to a hierarchy of schedulers. In this paper, we discuss why existing meta-scheduling hierarchies are sometimes not sufficient for Grid systems due to their inability to re-organise jobs already scheduled locally. Such a job re-organisation is required to adapt to evolving loads which are common in heavily used Grid infrastructures. We propose a peer-topeer scheduling model and evaluate it using case studies and mathematical modelling. We detail the DIANA (Data Intensive and Network Aware) scheduling algorithm and its queue management system for coping with the load distribution and for supporting bulk job scheduling. We demonstrate that such a system is beneficial for dynamic, distributed and self-organizing resource management and can assist in optimizing load or job distribution in complex Grid infrastructures.

High energy physics (HEP) and other scientific communities have adopted service oriented architectures (SOA) as part of a larger grid computing effort. This effort involves the integration of many legacy applications and programming libraries into a SOA framework. The grid analysis environment (GAE) (Lingen et al., 2004) is such a service oriented architecture based on the Clarens grid services framework (Steenberg et al., 2004) and is being developed as part of the compact muon solenoid (CMS) experiment at the large hadron collider (LHC) at European Laboratory for Particle Physics (CERN). Clarens provides a set of authorization, access control, and discovery services, as well as XMLRPC and SOAP access to all deployed services. Two implementations of the Clarens Web services framework (Python and Java) offer integration possibilities for a wide range of programming languages. This paper describes the Java implementation of the Clarens Web services framework called 'JClarens' and several Web services of interest to the scientific and grid community that have been deployed using JClarens.

In this paper we describe JClarens; a Java based implementation of the Clarens remote data server. JClarens provides Web services for an interactive analysis environment to dynamically access and analyze the tremendous amount of data scattered across various locations. Additionally this research is aimed to develop a service oriented grid enabled portal (GEP) that provides interface and access to several grid services to give a homogeneous and optimized view of the distributed and heterogeneous environment. Other than showing platform independent behavior provided by Java, the use of XML-RPC based Web services enabled JClarens to be a language neutral server and demonstrated interoperability with its Python variant. Extreme care has been taken in the usage and manipulation of various Java libraries to cater the needs of high performance computing. The overall exercise has yielded in a prototype with strong emphasis on security and virtual organization management (VOM). This shall provide a common platform to support development of larger, more flexible framework with future aims to integrate it with a loosely coupled, decentralized, and autonomous framework for grid enabled analysis environment (GAE).