Search

In the field of process mining, it is worth noting that process mining techniques assume that the resulting event logs can not only continuously record the occurrence of events but also contain all event data. However, like in IoT systems, data transmission may fail due to weak signal or resource competition, which causes the company’s information system to be unable to keep a complete event log. Based on a incomplete event log, the process model obtained by using existing process mining technologies is deviated from actual business process to a certain degree. In this paper, we propose a method for repairing missing activities based on succession relation of activities from event logs. We use an activity relation matrix to represent the event log and cluster it. The number of traces in the cluster is used as a measure of similarity calculation between incomplete traces and cluster results. Parallel activities in selecting pre-occurrence and post-occurrence activities of missing activities from incomplete traces are considered. Experimental results on real-life event logs show that our approach performs better than previous method in repairing missing activities.

Conselho Nacional de Desenvolvimento Cientifico e Tecnologico. Grant Number: 165009/2015-2 Fundacao de Amparo a Pesquisa do Estado de Sao Paulo (at CEMEAI). Grant Number: 2013/07375-0

Cloud computing provides an extensible infrastructure for executing workflows that demand high processing and storage capacity. Tasks are distributed and resources selected during scheduling where choices have a significant impact on data protection. Some workflow scheduling algorithms apply security services such as authentication, integrity verification, and encryption for both sensitive and non-sensitive tasks. However, this approach requires long makespan and monetary cost for execution. In this paper, we introduce a scheduling approach that considers the user annotation of workflow tasks according to the sensitiveness. We also optimize the scheduling using a multi-population genetic algorithm for minimizing cost while meeting a deadline. Extensive experiments using three workflow applications with different ratios of sensitive tasks and data size were performed to evaluate in terms of cost, makespan, risk, and wastage. The results showed that our approach can protect sensitive tasks more appropriately while achieving a better cost compared to other approaches in the literature.

Recommender systems are filters that suggest products of interest to customers, which may positively impact sales. Nowadays, there is a multitude of algorithms for recommender systems, and their performance varies widely. So it is crucial to choose the most suitable option given a situation, but it is not a trivial task. In this context, we propose the Recommender Systems Evaluator (RSE): a framework aimed to accomplish an offline performance evaluation of recommender systems. We argue that the usage of a proper methodology is crucial when evaluating the available options. However, it is frequently overlooked, leading to inconsistent results. To help appraisers draw reliable conclusions, RSE is based on statistical concepts and displays results intuitively. A comparative study of classical recommendation algorithms is presented as an evaluation, highlighting RSE’s critical features.

The microservice architecture is an architecture in which a single system is divided into small independently deployed services that are orchestrated together with the use of a lightweight mechanism. Each microservice does not rely much on other microservices (low coupling), but rather on its own resources to perform its task (high cohesion). This paper proposes a novel methodology which decomposes a monolith or other system into microservices in such a way that each microservice will function independently of other microservices while preserving some other key features, with the functionality distribution across each microservice being optimized with regards to usage of the functionality. This methodology makes use of dynamic analysis to identify resources that play a role in enabling the microservice to fulfill its functionality. We establish a set of rules which allows optimized distribution of the functionality. We evaluate the approach by applying it to real systems.

As research and development in Web Services and Services Computing has only existed for approximately a decade and half, it is remarkably that it has lead to a multi-billion dollar services industry. The first major research conference on the subject, the International Conference on Web Services, just started in 2003. Research in services computing is active and multifaceted. This special issue of IEEE Transactions on Services Computing considers two areas in services computing that have experienced “Recent Advances in Web Services Research”: Security, Privacy, and Trust; and New Approaches to Services Composition. It consists of the best extended papers from the two premier research conferences on services computing: the 2016 IEEE International Conference on Web Services and the 2016 IEEE Conference on Services Computing.

In this paper, four techniques of machine learning (ML) were applied and analyzed during the diagnosis of failures in vehicle fleet tracking modules. A comparison of the sampling methods was carried out considering the training and testing process using real data provided by DDMX, that acts in vehicle fleet tracking. A methodology was defined for pre-processing the collected data before the application of the ML techniques. Totally 16 models were created using the Random Forest (RF), Naive Bayes (NB), Support Vector Machine (SVM) and Multi Layer Perceptron (MLP) techniques. We have obtained promising results, where the techniques achieved a precision of 99.76% and 99.68% for detection and isolation of faults, respectively, on the provided dataset. These models can serve as prototypes to diagnose faults remotely and states that traditional ML techniques with manual feature extraction are still able to achieve high metrics.

Nowadays the access to a cloud computing environment is provided on-demand offering transparent services to customers. Although the cloud allows an abstraction of the behavior of the service providers in the infrastructure (involving logical and physical resources), it remains a challenge to fully comply with the Service Level Agreements (SLAs), because, depending on the service demand and system configuration, the providers may not be able to meet the requirements of the customers. There is a need for mechanisms that take account of load balancing algorithms to provide an efficient load distribution with the available resources. However, the studies in the literature do not effectively address the problem of the availability of resources to meet customers' requirements with analysis restricted to a limited set of objectives. This paper proposes algorithms to address the need for optimization when handling computational resources during the execution time. The methods optimizes the efficient use of the resources available in the infrastructure aiming to comply with the service level agreements defined between client and provider.

The growth of real‐world objects with embedded and globally networked sensors allows to consolidate the Internet of things paradigm and increase the number of applications in the domains of ubiquitous and context‐aware computing. The merging between cloud computing and Internet of things named cloud of things will be the key to handle thousands of sensors and their data. One of the main challenges in the cloud of things is context‐aware sensor search and selection. Typically, sensors require to be searched using two or more conflicting context properties. Most of the existing work uses some kind of multi‐criteria decision analysis to perform the sensor search and selection, but does not show any concern for the quality of the selection presented by these methods. In this paper, we analyse the behaviour of the SAW, TOPSIS and VIKOR multi‐objective decision methods and their quality of selection comparing them with the Pareto‐optimality solutions. The gathered results allow to analyse and compare these algorithms regarding their behaviour, the number of optimal solutions and redundancy.

The Internet of Things aims to digitize everyday physical objects by connecting them to the internet. As a result, cyber-physical environments of multiple sizes emerge, imposing new requirements on applications and software systems in regards support to heterogeneity and volatility. A challenging stage in the engineering of these systems is the validation. Although, there have been significant efforts to offer shared real-world testbeds, the simulations platforms are required to make the validation process cost and time effective. Existing simulation approaches only offer partial coverage to the key IoT environment characteristics, focus on communication or are specific for particular use cases and domains. In this paper, we propose a novel agent-based model that enables the simulation of the IoT systems with the key characteristics of an IoT environment. This model is designed to be flexible and adaptable to different experiments. Our approach introduces events in IoT environments as stochastic processes, enabling the evaluation of IoT systems under different conditions that otherwise would be time consuming and costly. We present the results of our experiments for evaluation of our model. These show that our proposal is a practical solution for the validation of IoT software systems, complementary to the real-world tests.

Scheduling is an important topic to support data security for workflow execution in clouds. Some workflow scheduling algorithms use security services such as authentication, integrity verification, and encryption for all workflow tasks. However, applying security services to no sensitive data does not make sense as no benefit is gained, yet it increases the makespan and monetary costs. In this paper, we introduce five policies for selection of tasks that handle sensitive data. We also propose a workflow scheduling algorithm based on a multi-populational genetic algorithm for minimizing cost while meeting a deadline. Experiments using four workflow applications show that our proposal can minimize both the makespan and cost, while maintaining the security of sensitive data compared to another approach in the literature.

In this study, a concept is presented that extends the My Remote previous project to a smart home solution, or the My Smart Remote. The concept extends its management capabilities for children's lives in the home, while at the same time safely and securely sharing collected data about children's activities with external parties including educationalists, health care professionals, psychologists and marketers of products and services used by children. Sharing data not only benefits these external parties but also the children themselves as solutions can be tailored to their needs. Because the My Smart Remote is for children, security and privacy concerns are paramount. A contribution of this concept that is primarily derived from the need for safety and security, and also efficiency in processing, is a solution where much of the sensor data of the My Smart Remote is processed locally, in addition to the cloud. This solution is based on the idea that smart solutions, such as smart cities and smart homes, generate a large amount of sensor data which can be more efficiently processed locally rather than depending on the cloud. The My Smart Remote using sensor data from around the home to manage television viewing, video gaming, activity on PCs, mobile phones and tablets, all in terms of time spent and types of activity. Additionally, the My Smart Remote will monitor these activities which will also include other activities such as taking drinks from the fridge and homework activities and collect, and share the associated data with external parties.

The Internet of Things (IoT) is able to connect billions of devices and services at anytime in any place, with various applications. Recently, the IoT became an emerging technology. One of the most significant current research discussion topics on the IoT is about the smart car parking. A modern urban city has over a million of cars on its roads but it does not have enough parking space. Moreover, most of the contemporary researchers propose management of the data on cloud. However, this method may be considered as an issue since the raw data is sent promptly from distributed sensors to the parking area via cloud and then received back after it is processed. This is considered as an expensive technique in terms of the data transmission as well as the energy cost and consumption. While the majority of proposed solutions address the problem of finding unoccupied parking space and ignore some other critical issues such as information about the nearest car parking and the roads traffic congestion, this paper goes beyond and proposes the alternative method. The paper proposes a smart car parking system that will assist users to solve the issue of finding a parking space and to minimise the time spent in searching for the nearest available car park. In addition, it provides users with roads traffic congestion status. Moreover, the proposed system collects the raw data locally and extracts features by applying data filtering and fusion techniques to reduce the transmitted data over the network. After that, the transformed data is sent to the cloud for processing and evaluating by using machine learning algorithms.

In a cloud computing environment, companies have the ability to allocate resources according to demand. However, there is a delay that may take minutes between the request for a new resource and it being ready for using. This causes the reactive techniques, which request a new resource only when the system reaches a certain load threshold, to be not suitable for the resource allocation process. To address this problem, it is necessary to predict requests that arrive at the system in the next period of time to allocate the necessary resources, before the system becomes overloaded. There are several time series forecasting models to calculate the workload predictions based on history of monitoring data. However, it is difficult to know which is the best time series forecasting model to be used in each case. The work becomes even more complicated when the user does not have much historical data to be analyzed. Most related work considers only single methods to evaluate the results of the forecast. Other works propose an approach that selects suitable forecasting methods for a given context. But in this case, it is necessary to have a significant amount of data to train the classifier. Moreover, the best solution may not be a specific model, but rather a combination of models. In this paper we propose an adaptive prediction method using genetic algorithms to combine time series forecasting models. Our method does not require a previous phase of training, because it constantly adapts the extent to which the data are coming. To evaluate our proposal, we use three logs extracted from real Web servers. The results show that our proposal often brings the best result and is generic enough to adapt to various types of time series.

The deluge of intelligent objects that are providing continuous access to data and services on one hand and the demand of developers and consumers to handle these data on the other hand require us to think about new communication paradigms and middleware. In hyper-scale systems, such as in the Internet of Things, large scale sensor networks or even mobile networks, one emerging requirement is to process, procure, and provide information with almost zero latency. This work is introducing new concepts for a middleware to enable fast communication by limiting information flow with filtering concepts using policy obligations and combining data processing techniques adopted from complex event processing.

Every day more and more objects are connected to the Internet to sense or actuate in some environment, composing the Internet of Things. IoT platforms will play a key role, as they will be responsible for managing low-level devices and data acquisition processes, and also support the development of new applications. One of the main challenges in IoT platforms will be the search and discovery of resources in large-scale and heterogeneous environments for reuse by other applications to support their specific requirements. In this paper, we propose an elimination-selection algorithm for search and discovery of resources in IoT environments. Our case study considers a real agricultural problem to be solved by the ViSIoT tool. The results show that our approach improves the quality of the proposed solution adding a small time overhead when compared to the TOPSIS algorithm used by ViSIoT.

Science Gateways have been widely accepted as an important tool in academic research, due to their flexibility, simple use and extension. However, such systems may yield performance traps that delay work progress and cause waste of resources or generation of poor scientific results. This paper addresses an investigation on some of the failures in a Galaxy system and analyses of their impacts. The use case is based on protein structure prediction experiments performed. A novel science gateway component is proposed towards the definition of the relation between general parameters and capacity of machines. The machine-learning strategies used appoint the best machine setup in a heterogeneous environment and the results show a complete overview of Galaxy, a diverse platform organization, and the workload behavior. A Support Vector Regression (SVR) model generated and based on a historic data-set provided an excellent learning module and proved a varied platform configuration is valuable as infrastructure in a science gateway. The results revealed the advantages of investing in local cluster infrastructures as a base for scientific experiments.

Scientific Gateways are one of the most important tools for designing and running experiments. Despite the possibility of local operation they are mainly available via online interfaces based on cloud computing instances. Our studies show that cloud machines may not be the best solution to every situation and that the advantages of heterogeneous cluster machines should be considered in scheduling experiments, saving both financial and computational resources, avoiding network delays and managing the infrastructure as needed. We run a variety of scenarios of bioinformatic experiments in three different sets of machines, a workstation, a cluster and cloud platform. Then, using Support Vector Machines (SVM), a nonlinear regression technique over the results, we can define the best machine configuration in terms of processing time according to the input parameters. The results show an approximation with a small error, that can define with good confidence the proper infrastructure to host a instance of the framework Galaxy, used as study case. With a system based on diverse environments, the researchers can properly schedule each set of experiments.

Nowadays, it is obvious that technology has revolutionised our lives by supporting us to do complicated jobs. The Internet of Things (IoT) is one of the emerging technologies. One of the most significant current research topics in the IoT is smart city. The smart city includes several applications such as smart home, smart industry and smart mobility. The smart car parking system is an aspect of smart mobility and an important application in smart city projects, because of the rapidly increasing number of cars in urban areas. However, most of the current proposals in smart car parking systems manage the data on the cloud side which is a problem since the system needs to send the raw data from sensor to cloud and receive instructions back: this is expensive in terms of energy and data transmission cost. To tackle this issue we present a proposal to save energy and to reduce the amount of data that is transmitted over the network to cloud by processing closer to source in this paper. The architecture is demonstrated through a case study.

This paper presents a cloud approach for low cost capacity planning evaluations. To perform these evaluations we have to specify and measure the workload on the target system to discover issues and make the necessary adjustments. However, due to high costs, these evaluations are usually done using simulations, which does not consider stochastic effects. We propose to use a tool named PEESOS, a generic and flexible approach to apply real workloads and measure used resources on these real systems. As a proof of concept, our case study use a real ticket sales service to evaluate the influence of scalability in the resource provisioning to show how PEESOS can lower the cost of such real evaluations. The results show the efficiency and savings that we can obtain using PEESOS for large-scale capacity planning evaluations before the real services are deployed. This approach can avoid several problems that real services faces when they launch.

IoT promises a world where Smart Objects (SO) are able to autonomously communicate and work together to make the (human) user's life easier. Popular approaches for development of IoT applications take for granted that on-object resources are evenly constrained. As consequence, we observe a trend towards a data-feeder architecture in which "smart objects" are simple data gatherers and senders. Raw data is stored and processed in cloud platforms feeding web applications and services. The autonomy of SOs is then compromised as they are not able to operate without these platforms. We propose a framework and architecture for the development of IoT applications where smart objects exhibit autonomy in regards to platforms and human users. We completed the successful evaluation of our proposal with the implementation of a prototype and the execution of a use case for physical resources provisioning.

The vast cloud computing environment holds out good prospects for researchers in the computing technology field. However, with several Cloud providers offering different pricing models, the evaluation and modeling of Cloud environments and applications are getting harder because there is a lack of tools for this task. We propose the CM Cloud Simulator to fill this gap since it provides a comprehensive and dynamic simulation of applications with various deployment configurations and incurs the cost it would require when implemented in a Cloud Provider, according to the cost model of any service provider. The CM Cloud Simulator also provides custom-built cost models through the XML file.

Traditional web services transaction processing mechanism handle exception by forward recovery and backward recovery. These compensation mechanisms often lead to waste of resources and time. In this paper, we propose a framework for predicting outcomes of service executions as part of service compositions which allows to choose service instances that are likely to lead to a successful result in the first instance and thus reduces the need for invoking costly recovery mechanisms. The framework makes use of watchdogs to maintain an awareness of service availability and a pre-coordinator which has oversight of the whole composite Web service and acts as a control center. An analysis of a scenario shows that we cannot only provide users with a more satisfactory result, but also can reduce the overhead costs of resources and waste.

Automatic Web services composition can be achieved using AI planning techniques. HTN planning has been adopted to handle the OWL-S Web service composition problem. However, existing composition methods based on HTN planning have not considered the choice of decompositions available to a problem, which can lead to a variety of valid solutions. In this paper, the authors propose a model of combining a Markov decision process model and HTN planning to address Web services composition. In the model, HTN planning is enhanced to decompose a task in multiple ways and find more than one plan, taking into account both functional and non-functional properties. Furthermore, an evaluation method to choose the optimal plan and experimental results illustrate that the proposed approach works effectively. The paper extends previous work by refining a number of aspects of the approach and applying it to a realistic case study.

Over the last few years, the number of smart objects connected to the Internet has grown exponentially in comparison to the number of services and applications. The integration between Cloud Computing and Internet of Things, named as Cloud of Things, plays a key role in managing the connected things, their data and services. One of the main challenges in Cloud of Things is the resource discovery of the smart objects and their reuse in different contexts. Most of the existent work uses some kind of multi-criteria decision analysis algorithm to perform the resource discovery, but do not evaluate the impact that the user constraints has in the final solution. In this paper, we analyse the behaviour of the SAW, TOPSIS and VIKOR multi-objective decision analyses algorithms and the impact of user constraints on them. We evaluated the quality of the proposed solutions using the Pareto-optimality concept., Comment: Proceedings of the 9th IEEE/ACM International Conference on Utility and Cloud Computing (UCC 2016) Shaghai, China, December, 2016