This paper introduces MarketSenseAI, an innovative framework leveraging GPT-4’s advanced reasoning for selecting stocks in financial markets. By integrating Chain of Thought and In-Context Learning, MarketSenseAI analyzes diverse data sources, including market trends, news, fundamentals, and macroeconomic factors, to emulate expert investment decision-making. The development, implementation, and validation of the framework are elaborately discussed, underscoring its capability to generate actionable and interpretable investment signals. A notable feature of this work is employing GPT-4 both as a predictive mechanism and signal evaluator, revealing the significant impact of the AI-generated explanations on signal accuracy, reliability, and acceptance. Through empirical testing on the competitive S&P 100 stocks over a 15-month period, MarketSenseAI demonstrated exceptional performance, delivering excess alpha of 10–30% and achieving a cumulative return of up to 72% over the period, while maintaining a risk profile comparable to the broader market. Our findings highlight the transformative potential of Large Language Models in financial decision-making, marking a significant leap in integrating generative AI into financial analytics and investment strategies.

Abstract Visual defect recognition and its manufacturing applications have been an upcoming topic in recent AI research. Defect datasets are often severely imbalanced and can be additionally burdened with separating classes of high visual similarity. Although various methods of data augmentation have been proposed to mitigate the class imbalance, they often fail to cope with tinier minority classes or have fidelity issues with smaller defects while, at the same time, needing significant computational resources to train. Also, augmentation based on vector-based oversampling struggles to produce high-fidelity inputs and is hard to apply on custom CNN architectures, which often perform better for this type of problem. Our work presents an image-level oversampling method based on an instance-based image generator that can be applied to any CNN directly during the training process without increasing the order of training time required. It is based on identifying a small number of the most uncertain base samples close to the estimated class boundaries and using them as seeds for augmentation. The resulting images are of high visual quality preserving small class differences, and they also improve the classifier boundary leading to higher recall scores than other state-of-the-art approaches.

Automated Machine Learning (AutoML) aims to identify the best-performing machine learning algorithm along with its input parameters for a given dataset and a specific machine learning task. This is a challenging problem, as the process of finding the best model and tuning it for a particular problem at hand is both time-consuming for a data scientist and computationally expensive. In this survey, we focus on unsupervised learning, and we turn our attention on AutoML methods for clustering. We present a systematic review that includes many recent research works for automated clustering. Furthermore, we provide a taxonomy for the classification of existing works, and we perform a qualitative comparison. As a result, this survey provides a comprehensive overview of the field of AutoML for clustering. Moreover, we identify open challenges for future research in this field.

In this paper, we present an architecture for mobility data spaces enabling trustworthy and reliable data operations along with its main constituent parts. The architecture makes use of a data lake for scalable storage of diverse mobility data sets, on top of which separate computing and storage layers are implemented to allow independent scaling with a data operations toolbox providing all data operations. Furthermore, to cater for mobility analytics, machine learning and artificial intelligence support, an edge analytics suite is provided that encompasses distributed algorithms for mobility analytics and federated learning, thereby exploiting edge computing technologies. In turn, this is supported by a resource allocator that monitors the energy consumption of data-intensive operations and provides this information to the platform for intelligent task placement in edge devices, aiming at energy-efficient operations. As a result, an end-to-end platform is proposed that combines data services and infrastructure services towards supporting mobility application domains, such as urban and maritime.

Serverless computing has reshaped the cloud computing landscape by offering benefits such as auto-scalability, streamlined operational management, and granular billing. As its adoption grows, challenges related to performance and cost optimization in hybrid architectures combining private servers and public cloud clusters have emerged. Central to these challenges are achieving optimal response latency and balancing performance and cost. To address these challenges, this paper introduces an adaptive routing service specifically designed for hybrid environments, proficient in leveraging real-time function metrics. Our proposed service pivots on three integral components: a monitor that captures performance metrics and raises alarms for predefined anomalies; a forecaster that predicts function latency across clusters, which includes wait and execution times and produces request distributions for each cluster to equalize the overall function latency; and a router then processes incoming requests, taking cues from the forecaster’s predictions. Notably, based on user-defined objectives, the forecaster can be directed to either minimize latency or optimize execution costs through trading off wait or execution time. Comprehensive evaluations on AWS and Azure clusters using the open source FaaS framework Apache OpenWhisk showcase our approach’s effectiveness, yielding a 9% improvement in average latency, a 45% decrease in standard deviation latency and a 17% cost reduction compared to conventional 50-50 routing. The advantages of elevated monitoring frequency are also illuminated, emphasizing quicker convergence times.

Text categorization and sentiment analysis are two of the most typical natural language processing tasks with various emerging applications implemented and utilized in different domains, such as health care and policy making. At the same time, the tremendous growth in the popularity and usage of social media, such as Twitter, has resulted on an immense increase in user-generated data, as mainly represented by the corresponding texts in users' posts. However, the analysis of these specific data and the extraction of actionable knowledge and added value out of them is a challenging task due to the domain diversity and the high multilingualism that characterizes these data. The latter highlights the emerging need for the implementation and utilization of domain-agnostic and multilingual solutions. To investigate a portion of these challenges this research work performs a comparative analysis of multilingual approaches for classifying both the sentiment and the text of an examined multilingual corpus. In this context, four multilingual BERT-based classifiers and a zero-shot classification approach are utilized and compared in terms of their accuracy and applicability in the classification of multilingual data. Their comparison has unveiled insightful outcomes and has a twofold interpretation. Multilingual BERT-based classifiers achieve high performances and transfer inference when trained and fine-tuned on multilingual data. While also the zero-shot approach presents a novel technique for creating multilingual solutions in a faster, more efficient, and scalable way. It can easily be fitted to new languages and new tasks while achieving relatively good results across many languages. However, when efficiency and scalability are less important than accuracy, it seems that this model, and zero-shot models in general, can not be compared to fine-tuned and trained multilingual BERT-based classifiers.

Edge computing has become a prominent solution when it comes to mobile applications and data management, due to its ability to considerably reduce data transmission costs, and to analyze data requiring fewer computing resources, since the analysis occurs at lower data volumes, without having to relocate data to centralized infrastructures. One major challenge, indicates the optimal data placement regarding data-intensive applications and, in general, applications requiring vast transmission of large amount of data. In this paper, we propose a novel user mobility-based data placement strategy, considering a trade-off between latency and data migration, which has not been investigated before. We classify the users into three mobility classes; namely static, local or mobile, via the use of a causal-aware Deep Learning network. This information is then exploited in order to optimize the data placement through specific data placement and retrieval algorithms for each mobility class. We evaluate the performance of the proposed solution using simulations, and prove that our solution manages to reduce the average data accessing cost by 60% for static or local users and 10% for mobile users, while the average path length is reduced by 50% for static and local users, and by 12% for mobile users.

Serverless computing has emerged as a revolutionary model that enables the deployment of applications and services by raising the level of abstraction from the underline resources. Its main functionality is enlightened by the notion of Function-as-a-Service (FaaS) as the core means to realize efficient serverless offerings. Following the shift from traditional architectures to microservices - by attaining flexibility, productivity, portability, and performance in industrial-scale IT projects, the serverless model introduces even more fine-grained services, named “nanoservices”, which facilitate required scalability by abstracting the deployment and management of the infrastructure resources. On the application space, advances in big data analysis contribute towards extracting actionable knowledge in various application domains. In this context, approaches for big data analysis aim at exploiting the added value of serverless architectures. To this end, we are presenting an extendable and generalized approach for facilitating the provision of Machine Learning Functions-as-a-Service (MLFaaS). The proposed approach outstrips the classical atomic and standard isolated services by facilitating composite services, i.e., workflows/pipelines of ML tasks, thus enabling the realization of the complete data path functions as required by data scientists. We demonstrate the operation of the proposed approach by modeling a real-world analytics scenario as an ML workflow pipeline and evaluate its performance in terms of performance. Furthermore, we address the challenge of utilizing a function oriented service template recommendation system, by expanding the serverless functional boundaries towards a holistic Quality-of-Service (QoS)-aware service function selection approach based on Artificial Intelligence techniques. These techniques propose the optimal number of functions to be implemented in a pipeline by exploiting the importance of response time as the primary key of the application’s performance.

Towards enabling the automated and optimized FaaS deployment of applications in a hybrid-cloud setting, the application requirements should be met by comparing them to the capabilities of the available resources of available clusters. To this end, semantic matching between the application characteristics and the individual descriptions of available compute clusters (e.g. from public or private cloud or edge facilities available) is required. In this work, such a system is proposed, namely the Reasoning Framework, which performs semantic matching between application and resource (meta)data and facilitates information sharing among the FaaS platform components leveraging Knowledge Graphs, ontology technologies, and semantic reasoning. The proposed system harvests information from the application function workflow, provided as a graph by the function editor specification (based on Node-RED), including developer-inserted annotations during the design process, and maps them to the dynamic information retrieved from the available clusters. The Reasoning Framework interprets these data as graphs and automatically applies several semantic rules that enable filtering of the available resources and efficient information retrieval through a RESTfull interface. The paper also discusses experimental results to further showcase the advantages of the proposed approach.

In recent years there is a surge of interest in the interpretability and explainability of AI systems, which is largely motivated by the need for ensuring the transparency and accountability of Artificial Intelligence (AI) operations, as well as by the need to minimize the cost and consequences of poor decisions. Another challenge that needs to be mentioned is the Cyber security attacks against AI infrastructures in manufacturing environments. This study examines eXplainable AI (XAI)-enhanced approaches against adversarial attacks for optimizing Cyber defense methods in manufacturing image classification tasks. The examined XAI methods were applied to an image classification task providing some insightful results regarding the utility of Local Interpretable Model-agnostic Explanations (LIME), Saliency maps, and the Gradient-weighted Class Activation Mapping (Grad-Cam) as methods to fortify a dataset against gradient evasion attacks. To this end, we “attacked” the XAI-enhanced Images and used them as input to the classifier to measure their robustness of it. Given the analyzed dataset, our research indicates that LIME-masked images are more robust to adversarial attacks. We additionally propose an Encoder-Decoder schema that timely predicts (decodes) the masked images, setting the proposed approach sufficient for a real-life problem.

The extraction of useful knowledge from collected data has always been the holy grail for enterprises and researchers, supporting efficient decision making, provided service's optimization and profit maximization. However, this task is easier said than done, since it presupposes the application of complex mathematical models/algorithms. Data Analysis has prospered due to the continuous demand to simplify and optimize the knowledge extraction process. Several mechanisms in different domains have been developed, consisting of various techniques to analyze specific data. The need for such mechanisms is even greater in healthcare, since there exist data of different complexity that may provide high-valuable knowledge, if properly analyzed. Considering these challenges, this paper proposes a mechanism for performing Data Analysis in diverse scenarios' healthcare data to extract valuable insights. The mechanism can collect data and apply several Machine Learning algorithms to ensure the best result about the prediction of certain features of the provided data.

The era of big data is surrounded by plenty of challenges, concerning aspects related to data quality, data management, and data analysis. Plenty of these challenges are met in several domains, such as the healthcare domain, where the corresponding healthcare platforms not only have to deal with managing and/or analyzing a tremendous quantity of health data, but also have to accomplish these actions in the most efficient and secure way possible. Towards this direction, medical institutions are paying attention to the replacement of traditional approaches such as the Monolithic and Service Oriented Architecture (SOA), which deal with many difficulties for handling the increasing amount of healthcare data. This paper presents a platform for overcoming these issues, by adopting the Microservice Architecture (MSA), being able to efficiently manage and analyze these vast amounts of data. More specifically, the proposed platform, namely beHEALTHIER, offers the ability to construct health policies out of data of collective knowledge, by utilizing a newly proposed kind of electronic health records (i.e., eXtended Health Records (XHRs)) and their corresponding networks, through the efficient analysis and management of ingested healthcare data. In order to achieve that, beHEALTHIER is architected based upon four (4) discrete and interacting pillars, namely the Data, the Information, the Knowledge and the Actions pillars. Since the proposed platform is based on MSA, it fully utilizes MSA's benefits, achieving fast response times and efficient mechanisms for healthcare data collection, processing, and analysis.

Multi-access Edge Computing (MEC) is expected to play an essential role in enabling 5G (and beyond) technologies and services. This has driven numerous micro-datacenter (μDC) deployment studies in the literature, with a common goal of addressing the optimal μDC placement and dimensioning problems. Along this line, this paper aims at clustering subareas with similar network activity dynamics, to find a good hotspots' representation over the urban area. Leveraging common Machine Learning (ML) and statistics principles, the main contribution of this paper is two-fold: (i) the definition and selection of dynamicity features based on real telecommunications datasets; and (ii) the granular cluster evaluation and analysis based on agglomerative hierarchical clustering. Three feature sets (containing 20, 12 and 8 features, respectively) are evaluated at varying precision levels, showing interesting trends on the number of clusters, heatmaps and intra-cluster correlation. These could potentially provide some valuable indications on the placement and dimensioning of the$\mu$DCs.

Simulations of Social Dynamics play an important role in public policy analysis and design as they help to estimate the effects that alternative policies can have on society. We argue for a tight integration of a novel generation of social dynamics meta-simulation tools in a policy development framework. Such tools provide a common, agent- based modeling and concurrent execution environment for creating and running such simulations. Furthermore, they enable reasoning about simulation outcomes by examining and ranking alternative policy objectives based on a set of criteria supplied by the designer. We describe our implementation of such an idea in Politika, our Elixir-based web framework for interactive, meta-simulation-based policy design and provide appropriate examples as part of the PolicyCLOUD EU research project.

Scientific and clinical research have advanced the ability of healthcare professionals to more precisely define diseases and classify patients into different groups based on their likelihood of responding to a given treatment, and on their future risks. However, a significant gap remains between the delivery of stratified healthcare and personalization. The latter implies solutions that seek to treat each citizen as a truly unique individual, as opposed to a member of a group with whom they share common risks or health-related characteristics. Personalisation also implies an approach that takes into account personal characteristics and conditions of individuals. This paper investigates how these desirable attributes can be developed and introduces a holistic environment, the iHELP, that incorporates big data management and Artificial Intelligence (AI) approaches to enable the realization of data-driven pathways where awareness, care and decision support is provided based on person-centric early risk prediction, prevention and intervention measures.

Automated machine learning (AutoML) aims to minimize human intervention during a machine learning task, for example by means of automatic algorithm selection and its configuration for the data set at hand. Although this research direction has attracted much interest lately, both in academia and industry, existing systems and tools mainly target the domain of supervised learning. However, unsupervised learning, in particular clustering, also calls for AutoML solutions, especially due to the ambiguity involved when evaluating clustering results. Motivated by this shortcoming, in this paper, we introduce a framework for automated clustering that encompasses two main modules: algorithm selection and hyperparameter tuning. Our approach to algorithm selection relies on meta-learning, based on novel meta-features extracted from data sets that attempt to capture similarities in the clustering structure. This approach is coupled with a method for hyperparameter tuning based on Bayesian optimization, where the main novelty is the proposal of an optimization goal that combines different cluster validity indices. We demonstrate the merits of our approach by empirical evaluation on 24 real-life data sets, which shows promising results when compared to existing methods.

Health Information Exchange (HIE) allows healthcare providers and citizens, to access and securely share healthcare information electronically, improving the speed, quality, safety, and cost of patient care. Exchange of this information can be achieved through a wired or wireless way, at close or long distances, achieving different goals in terms of transmission speed, exchange reliability and data transfer security. While HIE cannot replace provider-patient communication, it can greatly improve the completeness of patients' records. Most of the current research is devoted on exchanging health information among healthcare organizations, without giving the ability to the citizens on exchanging healthcare data with healthcare organizations and be able to manipulate this data, mainly due to lack of standardization and security guarantees. Towards the goal of HIE, and the ability of citizens to have access to their healthcare data, in this paper two different wireless communication protocols (Remote-to-Device (R2D) and Device-to-Device (D2D)) are specified that can be used by software applications. The goal of the R2D protocol is to facilitate the acquirement of healthcare data of a citizen from an Electronic Health Record (EHR) through internet connection, while the D2D protocol aims on facilitating the exchange of this health data among citizens and healthcare professionals, on top of Bluetooth.

We present the Disaster Recovery Layer (DRL) that enables OpenStack-managed datacenter workloads, Virtual Machines (VMs) and Volumes, to be protected and recovered in another datacenter, in case of a disaster. This work has been carried out in the context of the EU FP7 ORBIT project that develops technologies for enabling business continuity as a service. The DRL framework is based on a number of autonomous components and extensions of OpenStack modules, while its functionalities are available through OpenStack's Horizon UI and command line interface. Also, the DRL's architecture is extensible, allowing for the easy and dynamic integration of protection, restoration and orchestration plug-ins that adopt new approaches. A distributed disaster detection mechanism was also developed for identifying datacenter disasters and alerting the DRL. For the evaluation of the DRL, a two (active and backup) datacenters testbed has been setup in respective sites in Umea and Lulea, 265km apart and connected through the Swedish national research and education network. In case of a disaster, traffic is redirected between the datacenters utilizing the BGP anycast scheme. The experiments performed, show that DRL can efficiently protect VMs and Volumes, with minimum service disruption in case of failures and low overhead, even when the available bandwidth is limited.

The CrowdHEALTH project aims at delivering an integrated platform that provides decision support to public health authorities for policy creation through the exploitation of collective knowledge that emerges from multiple information sources. The latter will be realized through proposed Social Holistic Health Records - SHHRs. CrowdHEALTH will provide policy makers with the means of processing large amount of healthcare information (including diseases, root causes, risk factors and patient data across different geographical areas and timescales) from a single-entry point.

Big Data Analytics are indispensable components of architectures dealing with processing and visualizing results of diverse healthcare-related information sources. In this work, we propose a versatile cloud design where the Health Analytic Tools (HATs) are decoupled from the Datastore and the User-Interface parts, still preserving the element of system trust. This design offers advantages over the process of modifying and constructing new health policy models by means of supporting many-to-many relations between HATs and Health Key Performance Indicators. Additionally, it offers independence regarding HAT providers, analytics frameworks, cloud providers and deployment environments allowing the scaling of the proposed architecture.

The healthcare sector faces multiple challenges in implementing, maintaining and upgrading its systems, including technical, security and interoperability difficulties. Ontologies are the key to achieve and improve healthcare interoperability, in the form of a permanent artefact of the specification of each of the medical concepts. However, since different ontologies may have contradicting or overlapping parts, the need for the creation of ontology matching techniques for identifying similarities between them is rising. In the meantime, a large number of ontologies, vocabularies and taxonomies, as well as medical standards are constructed and globally adopted (e.g. HL7 FHIR), without promising the existence of a single global ontology, any time soon. Thus, a substantial overhaul of methodology is required to address the real complexity of health. This can be achieved through the FHIR Ontology Mapper (FOM) technique that is presented in this paper, which aims at constructing healthcare ontologies from any data source, and then identifying both the structural and semantic similarities between the latter and the HL7 FHIR resources ontologies. The final matchings derive from the mean of the latter, aiming at translating healthcare data of any type, into the widely adopted HL7 FHIR standard, thus enabling interoperability and improving the quality of patient care and research.

Interoperability between healthcare information systems is a challenge for the identification of important health-related events (through correlation of information across systems) and as a result for improving the patient care quality. Most of these systems are far from being interoperable and operate independently, whilst the value emerging from their exploitation is limited. While several techniques to address this challenge are based on medical standards and technologies, these techniques are not applicable to different scenarios, thus a holistic solution is needed. This paper focuses on the semantic interoperability of multiple electronic health records' (EHRs) - and their standards, proposing a multi-step generic semantic architecture that can be implemented simultaneously to different medical standards, for efficiently managing heterogeneous EHRs' data. The proposed architecture combines a mechanism for extracting Domain Specific information from classified EHRs' datasets, transforming them into a Common Health Language (CHL) through Ontologies, whereas unknown datasets are mapped and translated into CHL, using Ontology-Mapping techniques.

Cloud computing and virtualized infrastructures are currently the baseline environments for the provision of services in different application domains. While the number of service consumers increasingly grows, service providers aim at exploiting infrastructures that enable non-disruptive service provisioning, thus minimizing or even eliminating downtime. Nonetheless, to achieve the latter current approaches are either application-specific or cost inefficient, requiring the use of dedicated hardware. In this paper we present the reference architecture of a fault-tolerance scheme, which not only enhances cloud environments with the aforementioned capabilities but also achieves high-performance as required by mission critical every day applications. To realize the proposed approach, a new paradigm for memory and I/O externalization and consolidation is introduced, while current implementation references are also provided.

Cloud computing is becoming a widely adopted paradigm for the provision of different types of ICT services. Building on a set of combined technologies, it enables service provision following on-demand usage patterns. Different vendors aim at addressing the needs of private and public large and small organizations. To this end, several solutions have been developed tackling varying service, platform and infrastructure needs. In this paper, we present the outcomes of our experimentation with respect to two open-source infrastructure-level solutions, namely OpenStack and Synnefo which have attracted the attention both of the research community and of the industrial one. The goal of our qualitative and quantitative survey and experimentation is to identify shortcomings and areas for improvement in the widely adopted OpenStack project.

Cloud computing is a paradigm that aims to transform computer, storage and network resources into a utility. As more applications are deployed on cloud environments, one of the main requirements refers to efficient scalability during the application execution. In this paper we present two mechanisms that aim at addressing this requirement: an adaptable two layer monitoring mechanism and an event processing framework. We evaluate the effectiveness of these mechanisms through a set of experiments on a large scale multi-cloud facility. The latter poses challenges with respect to time-constrained execution of applications, since the aforementioned mechanisms need to collect and analyse information from geographically distributed sites, and trigger scaling decisions during the real-time application execution. The experimentation outcomes demonstrate the value of the presented mechanisms, both in terms of efficient scalability and with respect to the introduced overhead on the infrastructure.

Cloud services have emerged as an innovative IT provisioning model in the recent years. However, after their usage severe considerations have emerged with regard to their varying performance due to multitenancy and resource sharing issues. These issues make it very difficult to provide any kind of performance estimation during application design or deployment time. The aim of this paper is to present a mechanism and process for measuring the performance of various Cloud services and describing this information in machine understandable format. The framework is responsible for organizing the execution and can support multiple Cloud providers. Furthermore we present approaches for measuring service performance with the usage of specialized metrics for ranking the services according to a weighted combination of cost, performance and workload.

Recently, the idea that the Internet of Things (IoT) systems can be advantaged in many ways by integrating social networking concepts is gaining momentum. In this paper we present the social approach that the COSMOS project introduces. COSMOS supports knowledge flow between Things in order to provide a system that learns, observes and evaluates the usage and communication patterns and generates new knowledge. It focuses on the value of experience and experience-sharing and investigates models and principles designed for the social networks, which would provide it with the potential to support novel applications in more effective and efficient ways.

Currently there is significant and increasing demand for quick data storage and access any time. Storage is a fundamental need and Cloud environments accomplish this. The evolvement of the IT technologies offers the user to a large variety of useful services. In order for the user to have a guarantee for these services SLAs are signed. In this paper, we present an SLA Management mechanism for Cloud environments. SLA enforcement is based on rules that are updated in runtime in order to proactively detect possible SLA violations and handle them in an appropriate manner.

Future Internet applications raise the need for infrastructures that can facilitate real-time and interactivity without major modifications in the application domain. Such infrastructures should be able to efficiently adapt resource provisioning to the dynamic demands of the applications, the majority of which tends to be real-time and interactive, posing specific Quality of Service (QoS) requirements. In this context, we present initial experimentation outcomes of a set of mechanisms (namely QoS-oriented Service Management mechanisms) that aim at enabling the provision of QoS guarantees in cloud infrastructures. They incorporate an orchestration framework that allows for workflow management in the case of non-monolithic applications, an adaptable monitoring mechanism that collects and aggregates monitoring information from application and infrastructure levels, and an evaluator service enabling runtime adaptability through provisioning policies. We also demonstrate the operation of the implemented mechanisms and evaluate their effectiveness in a large-scale multi-site cloud facility.

In a world of multi-stakeholder information and assets provision on top of millions of real-time interacting and communicating things, systems based on Internet of Things (IoT) technologies aim at exploiting these assets in a resilient and sustainable way allowing them to reach their full potential. In this paper we present two innovative smart city IoT applications: the first one refers to heat and energy management, and aims at utilizing different resources (such as heat and electricity meters) in order to optimize use of energy in commercial and residential areas. The second application refers to cruise control for public transportation, and aims at utilizing different resources (such as environmental and traffic sensors) in order to provide driving recommendations that aim at eco efficiency. We also highlight the IoT challenges as well as potential enabling technologies that will allow for the realization of the proposed applications.

In modern utility computing infrastructures, like grids and clouds, one of the significant actions of a service provider is to predict the resources needed by the services included in its platform in an automated fashion for service provisioning optimization. Furthermore, a variety of software toolkits exist that implement an extended set of algorithms applicable to workload forecasting. However, their automated use as services in the distributed computing paradigm includes a number of design and implementation challenges. In this paper, a decoupled framework is presented, for taking advantage of software like GNU Octave in the process of creating and using prediction models during the service life cycle of a SOI. A performance analysis of the framework is also conducted. In this context, a methodology for creating parametric or gearbox services with multiple modes of operations based on the execution conditions is portrayed and is applied to transform the aforementioned service framework to optimize service performance. A new estimation algorithm is introduced, that creates performance rules of applications as black boxes, through the creation and usage of genetically optimized artificial neural networks. Through this combination, the critical parameters of the networks are decided through an evolutionary iterative process.

The emergence of Cloud technologies ultimately affected the service computing ecosystem introducing new roles and relationships as well as new architectural and business models. Along with the increase of the capabilities and potentials of the service providers came the increase of the information being available and the need to manage it in an efficient way. To this end, in this paper we present a management service architecture as well as the information model that the solution is based on. The model has been designed to serve a storage service provided by a Cloud infrastructure but the approach is implemented in a flexible and modular fashion in order to support different Cloud situations.

In this paper, we address the problem of managing SLAs in cloud computing environments. The idea is to take advantage of the content terms that concern the objects and support more efficient capabilities, such as quicker search and retrieval of the objects. As a result, the operational cost is reduced and consequently this fact lessens the customer's charge.

The emergence of cloud environments has made feasible the delivery of Internet-scale services by addressing a number of challenges such as live migration, fault tolerance and quality of service. However, current approaches do not tackle key issues related to cloud storage, which are of increasing importance given the enormous amount of data being produced in today's rich digital environment (e.g. by smart phones, social networks, sensors, user generated content). In this paper we present the architecture of a scalable and flexible cloud environment addressing the challenge of providing data-intensive storage cloud services through raising the abstraction level of storage, enabling data mobility across providers, allowing computational and content-centric access to storage and deploying new data-oriented mechanisms for QoS and security guarantees. We also demonstrate the added value and effectiveness of the proposed architecture through two real-life application scenarios from the healthcare and media domains.

The emergence of new environments such as Cloud computing highlighted new challenges in traditional fields like performance estimation. Most of the current cloud environments follow the Software, Platform, Infrastructure service model in order to map discrete roles / providers according to the offering in each “layer”. However, the limited amount of information passed from one layer to the other has raised the level of difficulty in translating user-understandable application terms from the Software layer to resource specific attributes, which can be used to manage resources in the Platform and Infrastructure layers. In this paper, a generic black box approach, based on Artificial Neural Networks is used in order to perform the aforementioned translation. The efficiency of the approach is presented and validated through different application scenarios (namely FFMPEG encoding and real-time interactive e-Learning) that highlight its applicability even in cases where accurate performance estimation is critical, as in cloud environments aiming to facilitate real-time and interactivity.

The advancements in distributed computing have driven the emergence of service-based infrastructures that allow for on-demand provision of ICT assets. Taking into consideration the complexity of distributed environments, significant challenges exist in providing and managing the offered on-demand resources with the required level of Quality of Service (QoS), especially for real-time interactive multimedia applications. Common to these applications is the service-oriented approach being followed, thus applications consist of application service components that interact in order to provide the corresponding application functionality. In this paper we present an architectural design of a complete Workflow Management System (WfMS) for enacting application service components that have been deployed in a Virtualized Environment. The WfMS aims at synchronizing the application components, monitoring their execution and handling faults while adhering to the QoS requirements of the application.

This paper deals with the problem of scheduling workflow applications with quality of service (QoS) constraints, comprising real-time and interactivity constraints, over a service-oriented grid network. A novel approach is proposed, in which high-level advance reservations, supporting flexible start and end time, are combined with low-level soft real-time scheduling, allowing for the concurrent deployment of multiple services on the same host while fulfilling their QoS requirements. By undertaking a stochastic approach, in which a-priori knowledge is leveraged about the probability of activation of the application workflows within the reserved time-frame, the proposed methodology allows for the achievement of various trade-offs between the need for respecting QoS constraints (user perspective) and the need for having good resource saturation levels (service provider perspective).

Abstract The advent of heterogeneous and distributed environments, such as Grid environments, made feasible the solution to computational‐intensive problems in a reliable and cost‐effective manner. In parallel, workflows with increased complexity that require specialized systems to deal with them are emerging, so as to carry out more composite and mission‐critical applications. In that rationale, quality‐of‐service (QoS) issues need to be tackled in order to ensure that each application satisfies the corresponding user requirements. Therefore, considering the quality provision aspect as fundamental for enabling Grid applications to become QoS compliant, we present an approach for service selection using QoS criteria. The latter is achieved with a suite of components that allow the different mappings of application workflow processes to Grid services that not only meet the user goals and requirements but also maximize his/her benefit in terms of the offered QoS level. We also demonstrate the operation of the aforementioned suite of components and evaluate its performance and effectiveness using a Grid scenario, based on a 3D image rendering application. Copyright © 2008 John Wiley & Sons, Ltd.

Abstract Grids and mobile Grids can form the basis and the enabling technology for pervasive and utility computing due to their ability to being open, highly heterogeneous and scalable. In this paper we present a scheme for advancing quality of service (QoS) attributes, such as fault tolerance and prioritized scheduling, in OGSA‐based mobile Grids. The fault tolerance is achieved by producing and managing sufficient replicas of tasks submitted for execution on the mobile Grid resources. We design a simple and efficient prioritization scheme, which allows the scheduling of the tasks submitted by the Grid users as distinguished priorities that can be managed and exploited as a QoS parameter by the Grid infrastructure operator. The results that are presented show the efficiency of the proposed scheme in being simple and additionally enriching with reliability and QoS features the applications that are built on the concept of mobile Grids. Copyright © 2008 John Wiley & Sons, Ltd.

A constantly increasing number of applications from various scientific sectors are finding their way towards adopting Grid technologies in order to take advantage of their capabilities: the advent of Grid environments made feasible the solution of computational intensive problems in a reliable and cost-effective way. The aim of this paper is to demonstrate how multilevel tumour growth and response to therapeutic treatment models can be used in order to simulate clinical trials, with the long-term intention of better designing clinical studies and understanding their outcome based on basic biological science. For this purpose, a computer simulation model of glioblastoma multiforme response to radiotherapy has been applied to perform the aforementioned simulation in a real Grid environment by also taking into account historical data. The proposed approach yields very good results for the conducted virtual trial since these are in agreement with the outcome of the real clinical study, while the use of Grid technologies demonstrate and highlight their added-value.

This paper advocates the need for a mechanism that will allow the evaluation of the provided quality of service (QoS) by a service provider to a service customer in B2B service provisioning. Furthermore, this study goes on with presenting the feasibility of this mechanism by designing and testing a reference implementation that can be used in a service oriented architecture (SOA) environment which is able to support business application services. Experience gained in the frame of the NextGRID IST project that focused on the business perspectives of Grid computing and adopted SOA as its baseline architecture, has shown that such a mechanism is essential for enabling the economic viability of such large scale computing platforms.

A constantly increasing number of applications from various scientific sectors are finding their way towards adopting grid technologies in order to take advantage of their capabilities: the advent of grid environments made feasible the solution of computational intensive problems in a reliable and cost-effective way. In this paper we present a grid-based approach for aggregation of data that are obtained from various sources (e.g. cameras, sensors) and their analysis with the use of genetic algorithms. By also taking into consideration general historical data and patient-specific medical information, we present the realization of the proposed approach with an application scenario for personalized healthcare and medicine.

Abstract Enabling e‐Business applications on Grid computing has been an important topic of research as the beneficial perspective of Grids lies on solving computational intensive problems by large‐scale distributed resources. In this paper we present a toolkit for centralized workload prediction and access, in order to meet the requirements of execution of commercial business processes on Grid infrastructures as well as to take advantage of their cost‐effective capability for industrial users. The implementation of centralized workload prediction is achieved with an artificial neural network—available in the toolkit—each time a user completes a job execution. This approach has been adopted and validated within the framework of a GRIA IST project for aspecific industrial application, namely 3D image rendering, which meets the e‐Business application's standards. The accuracy of the prediction showed promising results and in combination with the centralized access capability makes e‐Business applications feasible on Grids. Copyright © 2006 John Wiley & Sons, Ltd.