Smart Infrastructure and Applications

Road traffic congestion is one of the most significant problems in the world, especially in large cities. In Saudi Arabia, accidents and traffic jams have increased in many major roads due to the lack of public transportation, increasing number of vehicles, and an enormous number of pilgrim visitors all year round. Twitter has emerged as an important source of information on various topics including road traffic. A large number of tweets are posted every day by users who wish to inform their followers about traffic conditions. Moreover, big data processing technologies provide unprecedented data analysis opportunities for addressing transportation problems. In this paper, we introduce a methodology for preprocessing and analyzing traffic-related tweets in the Arabic language, particularly the Saudi dialect using a big data processing platform (SAP HANA). Furthermore, we propose a technique for sentiment classification using lexicon-based approach to understand driver’s feelings. We collect tweets from Jeddah and Makkah cities and identify the most congested roads in the cities. We also detect events of multiple types: accidents, roadworks, fire, weather conditions, and others. The causes for the congestion in the cities are also identified.

High performance computing (HPC), big data, and artificial intelligence (AI) technologies are playing a key role in enabling smart society systems to sense the cities and other environments at micro-levels, detecting events, making intelligent decisions, and taking appropriate actions, all within stringent time bounds. Social media have revolutionized our societies and are gradually becoming a key pulse of smart societies by sensing the information about the people and their spatio-temporal experiences around the living spaces. The aim of this work is to develop data management and analysis techniques for smart societies. Specifically, we use big data, machine learning, and other platforms including Spark, MLlib, Tableau, and Google Maps Geocoding API, to study Twitter data for the detection and validation of spatio-temporal events in London. We empirically demonstrate that physical, virtual, and conceptual events can be detected automatically by analyzing data. We find and locate congestion around London. We detect the occurrence of multiple events including “Underbelly Festival,” “The Luna Cinema” and “London Notting Hill Carnival 2017,” their locations and times, without any prior knowledge of the events. An architecture of our big data analytics tool based on Spark for the detection of spatio-temporal events is provided along with the details of the main system components using six algorithms. The event detection pipeline has been enhanced using a methodology to automatically validate the factuality of the detected events. We also provide a comparison of three machine learning methods, support vector machine, logistic regression, and Naïve Bayes for event detection.

A staggering 1.25 million people die and up to 50 million people suffer injuries annually due to road traffic crashes around the world, causing great socio-economic and environmental damages. Road collisions are a major cause of road congestion. The cost of congestion to the US economy, alone, exceeded 305 billion USD in 2017. Smart infrastructure developments have accelerated the pace of technological advancements and the penetration of these technologies to all spheres of everyday life including transportation. The use of GPS devices to collect data, image processing and artificial intelligence (AI) for traffic analysis, and autonomous driving are but a few examples. This paper brings together transport big data, deep learning, in-memory computing, and GPU computing to predict traffic incidents on the road. Three different kinds of datasets—road traffic, vehicle detector station (VDS), and incident data—are combined together to predict road traffic incidents. The data is acquired from the California Department of Transportation (Caltrans) Performance Measurement System (PeMS). We have analyzed over 10 years of road traffic data. This work-in-progress paper reports incident prediction results using 3 months’ data, September to November 2017. The data fusion methodology is explained in detail along with the algorithms. The results for various configurations of deep convolution neural networks are given. Conclusions are drawn from the current status of the results and ideas for future improvements are given.

Mobility is one of the major dimensions of smart city design and development. Transportation analysis and prediction play important parts in mobility research and development. Recent years have seen many new types of transportation data emerging, such as social media and Global Positioning System (GPS) data. These data contains hidden knowledge, which can be used in many applications to improve city operations; road traffic prediction is one aspect of this. Researchers have traditionally used single traffic flow prediction methods, which work well only under specific conditions. Some work has emerged in recent years on combining these methods into various hybrid methods. However, this work is in its infancy, and further investigations are required. More importantly, these hybrid methods have mostly been developed on stand-alone, nondistributed platforms, limiting the data and problem sizes that can be addressed, as well as the accuracy that can be achieved. This chapter gives a review of traffic flow prediction and modeling methods and discusses the limitations of each method. A review of the various types of transportation traffic data sources is provided. Notable big data analysis tools, including the Apache Spark platform, are described. Finally, we describe a hybrid method for road traffic prediction and provide a tutorial on the process of hybrid traffic flow prediction. The hybrid method is based on the autoregressive integrated moving average (ARIMA) and support vector machine (SVM) methods.

Road transportation is among the grand global challenges affecting human lives, health, society, and economy. Autonomous vehicles (AVs) are the latest among the radical solutions to address transportation challenges. AVs have become a reality although their penetration in real environments needs more time. The foremost challenge for AVs is to recognize objects in real driving environment with highest certainty. This paper is an extension of our earlier work where we developed a methodology to integrate supervised learning and decision fusion to enhance object classification accuracy in a driving environment, i.e., to enable an auto-pilot to take better driving decisions. This problem equates to pixel classification. Our study revealed that the C5.0 decision tree classifier performs similar to deep learning. This paper extends and investigates the topic further and provides an in-depth performance comparison of deep learning and C5.0 decision tree classifier for object classification in driving environments using a bigger dataset. We manually label images from a subset of KITTI road dataset by using free-form selection (polygon) rather than a box or rectangular selection enabling highly accurate pixel labeling. Our analysis reveals that C5.0 and deep learning provide similar accuracies while deep learning is over 30% faster than C5.0.

Natural and manmade disasters have increased significantly over the past few decades. These include, among many others, the recent floods in Japan (June/July 2018) and Barcelona attack of August 2017. The Japan floods had left around 200 people dead, 70 were reported missing, and over eight million people were ordered to evacuate their homes, with 2bn USD the estimated cost of flood rebuilding. Disaster management plays a key role in reducing the human and economic losses. Our earlier work has focused on developing a disaster management system leveraging technologies including vehicular ad hoc networks (VANETs) and cloud computing to devise city evacuation strategies. The work was later extended to incorporate traffic management plans for smart cities using deep learning techniques. In-memory computations and graphics processing units (GPUs) were used to address intensive and timely computational demands of deep learning over big data in disaster situations. This paper extends our earlier work and provides extended analysis and results of the proposed system. A system architecture based on in-memory big data management and GPU-based deep learning computations is proposed. We have used road traffic data made publicly available by the UK Department for Transport. The results show the effectiveness of the deep learning approach in predicting traffic behavior in disaster and evacuation situations. This is the first system which brings together deep learning, in-memory data-driven computations, and GPU technologies for disaster management.

This chapter reports our continuing work on single source shortest path computations of big data road network graphs using Apache Spark. Smart applications and infrastructures are increasingly relying on graph computations to model real-life problems. Big data is being generated from various sources such as Internet of Things (IoT) and social media. Big data cannot be processed by traditional tools and technologies due to their properties, volume, velocity, veracity, and variety. The problems and relevant data are typically large and, hence, give rise to large graphs, which could be analyzed and solved using big data technologies. We use the US road network data, modelled as graphs, and calculate shortest paths between a set of large numbers of vertices in parallel. The experiments are performed on the Aziz supercomputer. We analyze Spark’s parallelization behavior by solving problems of varying graph sizes, i.e., various states of the USA (with over 58 million edges), and varying number of shortest path queries up to one million. We achieve good performance, and as expected, the speedup is dependent on both the size of the data and the number of parallel nodes. The system architecture for graph computing in Spark is explained. A detailed review of the relevant work is provided. We call our system, the Big Data Shortest Path Graph Computing (BDSPG) system.

DNA typing or profiling is being widely used for criminal identification, paternity tests, and diagnosis of genetic diseases. DNA typing is considered one of the hardest problems in the forensic science domain, and it is an active area of research. The computational complexity of DNA typing increases significantly with the number of unknowns in the mixture and has been the major deterring factor holding its advancements and applications. In this chapter, we provide an extended review of DNA profiling methods and tools with a particular focus on their computational performance and accuracy. The process of DNA profiling within the broader context of forensic science and genetics is explained. The various classes of DNA profiling methods including general methods, and those based on maximum likelihood estimators, are reviewed. The reviewed DNA profiling tools include LRmix Studio, TrueAllele, DNAMIX V.3, Euroformix, CeesIt, NOCIt, DNAMixture, Kongoh, LikeLTD, LabRetriever, and STRmix. A review of high-performance computing literature in bioinformatics and HPC frameworks is also given. Faster interpretations of DNA mixtures with a large number of unknowns and higher accuracies are expected to open up new frontiers for this area.

Technology plays a vast role in all aspects of our lives. In every business, technology is helpful to fulfill the needs of the customers. Cloud computing is a technology that satisfies the demand for dynamic resources and makes it easier for jobs to work on all platforms. Cloud computing platforms on the internet provide rapid access with pay-as-you-go pricing. Many business organizations have deployed their data either fully or partially on a cloud platform. A healthcare cloud is a platform where one can easily find hospitals, doctors, clinics, pharmacies, etc. While one face of this cloud is quite beneficial, the other face is challenging because of security issues. The data are stored somewhere else; hence, they can be an attractive target for cybercriminals. This chapter provides an introduction to cloud computing and the healthcare cloud. Subsequently, security issues in cloud computing, especially in the context of the healthcare cloud, are introduced. Finally, some methods to improve cloud security for healthcare are discussed along with our proposed architecture.

It is estimated that healthcare spending in the world’s major regions will increase from 2.4% of GDP to 7.5% during 2015 to 2020. Healthcare providers are required to deliver high-quality medical services to their customers. Since most of their budgets are spent on high cost medical equipment and medicines, there is a pressing need for them to optimize their supply chain activities such that high-quality services could be provided at lower costs. Relatedly, medical equipment and devices generate massive amounts of unused data. Big data analytics is proven to be helpful in forecasting and decision-making, and, hence, can be a powerful tool to improve healthcare supply chains. This paper extends our earlier work and presents a review on the use of big data in healthcare supply chains. We review the various concepts related to the topic of this paper including big data, big data analytics, and the role of big data in healthcare, supply chain management (SCM) and healthcare supply chain management. The role of Twitter data in SCM is also explored. The opportunities and challenges for big data enabled healthcare supply chains are discussed along with several directions for future developments. We conclude that the use of big data in healthcare supply chains is of immense potential and demands further investigation.

Smart city systems are fast emerging as solutions that provide better and digitized urban services to empower individuals and organizations. Mobile and cloud computing technologies can enable smart city systems to (1) exploit the portability and context-awareness of mobile devices and (2) utilize the computation and storage services of cloud servers. Despite the wide-spread adoption of mobile and cloud computing technologies, there is still a lack of solutions that provide the users with portable and context-aware recommendations based on their localized context. We propose to advance the state-of-the-art on recommender systems—providing a portable, efficient, and context-driven digital matchmaking—in the context of smart markets that involves virtualized customers and business entities. We have proposed a framework and algorithms that unify the mobile and cloud computing technologies to offer context-aware and portable recommendations for smart markets. We have developed a prototype as a proof-of-the-concept to support automation, user intervention, and customization of users’ preferences during the recommendation process. The evaluation results suggest that the framework (1) has a high accuracy for context-aware recommendations, and (2) it supports computation and energy efficient mobile computing. The proposed solution aims to advance the research on recommender systems for smart city systems by providing context-aware and portable computing for smart markets.

Data mining (DM) is gaining significant importance these days because of the flow and accumulation of data from various sources and fields; it has been estimated that the world’s databases double every 20 months (Tan et al. Introduction to data mining, Pearson Addison Wesley, Boston, 2005. The data are growing not only in terms of their volume but also in terms of their complexity and diversity. Thus, DM techniques and tools are required, and many algorithms and techniques have been introduced. These techniques include clustering, association, and prediction via classification and regression. These techniques are used in many fields such as business, healthcare, and education. In this chapter, we explore the application of some of these techniques in education. In the educational field, instructors tend to use their experience and personal judgment to link grades and failures of students between courses on the basis of their knowledge of the courses’ content. As a result, major plans may be changed, and academic guidance is offered accordingly. However, such an opinion is not always validated and tested, and we cannot be certain of it. With the existence of DM techniques, along with the vast volumes of data held by education systems, universities and schools can predict students’ performance and find associations between many attributes such as course grades. The results of using these techniques can have a profound effect in helping to change program plans and offer guidance. Students can make better-informed decisions when presented with facts that can have effects on their study. In this chapter, we review DM techniques used to mine student data with a focus on association rule mining. We report our work on association rule mining of Computer Science students’ grades and address some related issues. In this work we used lift, Kulczynski (Kulc), and the imbalance ratio (IR) to measures the interestingness of the rules. Our results showed cases of correlation between courses with confidence from 80 to 100%.

Web applications and services are fundamental to designing smart infrastructure and cities. Developers often use various development technologies when developing web or cloud applications. One of such major technologies is web frameworks (e.g., Rails, Spring, Django, and CodeIgniter), which permit developers to develop without worrying about the low-level details. Programmers may choose from a variety of web frameworks, and different languages that support them, each with its own strengths and weaknesses. Organizations work in different application domains and have diverse priorities and constraints with regard to the development of applications and services. In this paper, we propose an automatic tool, SelecWeb, for selecting a web framework based on a set of criteria and developer preferences. The set of selection criteria is developed by us and is a contribution of this paper. The tool currently uses analytic hierarchy process (AHP) for comparison, analysis, and decision-making. We provide a detailed description and analysis of the tool including a case study for web framework selection. Conclusions are drawn with design ideas on future extension of the tool using machine learning.

In recent times, the commodity Single Board Computers (SBCs) have now become sufficiently powerful that they can run standard operating systems and mainstream workloads. In this chapter, we investigate the design and implementation of Single Board Computer (SBC)-based Hadoop clusters. We provide a compact design layout and build two clusters each using 20 nodes. We extensively test the performance of these clusters using popular performance benchmarks for task execution time, memory/storage utilization, network throughput, and energy consumption. We investigate the cost of operating SBC-based clusters by correlating energy utilization for the execution time of various benchmarks using workloads of different sizes. Although the low-cost benefit of a cluster built with ARM-based SBCs is desirable, these clusters yield low comparable performance and energy efficiency due to limited onboard capabilities. It is, however, possible to tweak Hadoop configuration parameters to ARM-based SBC specifications to efficiently utilize available resources. Finally, a discussion on the design implications of these clusters as a testbed for inexpensive and green cloud computing research is presented.

Many important scientific, engineering, and smart city applications require solving large sparse linear equation systems. The numerical methods for solving linear equations can be categorised into direct methods and iterative methods. Jacobi method is one of the iterative solvers that has been widely used due to its simplicity and efficiency. Its performance is affected by factors including the storage format, the specific computational algorithm, and its implementation. While the performance of Jacobi has been studied extensively on conventional CPU architectures, research on its performance on emerging architectures, such as the Intel Many Integrated Core (MIC) architecture, is still in its infancy. In this chapter, we investigate the performance of parallel implementations of the Jacobi method on Knights Corner (KNC), the first generation of the Intel MIC architectures. We implement Jacobi with two storage formats, Compressed Sparse Row (CSR) and Modified Sparse Row (MSR), and measure their performance in terms of execution time, offloading time, and speedup. We report results of sparse matrices with over 28 million rows and 640 million non-zero elements acquired from 13 diverse application domains. The experimental results show that our Jacobi parallel implementation on MIC achieves speedups of up to 27.75× compared to the sequential implementation. It also delivers a speedup of up to 3.81× compared to a powerful node comprising 24 cores in two Intel Xeon E5-2695v2 processors.

The massive parallelism provided by the graphics processing units (GPUs) offers tremendous performance in many high-performance computing applications. One such application is Sparse Matrix-Vector (SpMV) multiplication, which is an essential building block for numerous scientific and engineering applications. Researchers who propose new storage techniques for sparse matrix-vector multiplication focus mainly on a single evaluation metrics or performance characteristics which is usually the throughput performance of sparse matrix-vector multiplication in FLOPS. However, such an evaluation does not provide a deeper insight nor allow to compare new SpMV techniques with their competitors directly. In this chapter, we explain the notable performance characteristics of the GPU architectures and SpMV computations. We discuss various strategies to improve the performance of SpMV on GPUs. We also discuss a few performance criteria that are usually overlooked by the researchers during the evaluation process. We also analyze various well-known schemes such as COO, CSR, ELL, DIA, HYB, and CSR5 using the discussed performance characteristics.

High-performance computing (HPC) plays a key role in driving innovations in health, economics, energy, transport, networks, and other smart-society infrastructures. HPC enables large-scale simulations and processing of big data related to smart societies to optimize their services. Driving high efficiency from shared-memory and distributed HPC systems have always been challenging; it has become essential as we move towards the exascale computing era. Therefore, the evaluation, analysis, and optimization of HPC applications and systems to improve HPC performance on various platforms are of paramount importance. This paper reviews the performance analysis tools and techniques for HPC applications and systems. Common HPC applications used by the researchers and HPC benchmarking suites are discussed. A qualitative comparison of various tools used for the performance analysis of HPC applications is provided. Conclusions are drawn with future research directions.

The outburst of data produced over the last few years in various fields has demanded new processing techniques, novel big data–processing architectures, and intelligent algorithms for effective and efficient exploitation of huge data sets to get useful insights and improved knowledge discovery. The explosion of data brings many challenges to deal with the complexity of information overload. Numerous tools and techniques have been developed over the years to deal with big data challenges. This chapter presents a summary of state-of-the-art tools and techniques for processing of big data applications by critically analyzing their objectives, methodologies, and key approaches to address the challenges associated with big data. Also, we critically analyze some of the core applications of big data and their impacts in improving the quality of human life by primarily focusing on healthcare and smart city applications, genome sequence annotation applications, and graph-based applications. We provide a detailed review and taxonomy of the research efforts within each application domain.

Big data is being at the forefront of many ICT-based developments in all spheres of life, be it business, education, or entertainment. Big data is being generated from many diverse sources including social media, Internet of Things (IoT), manufacturing and operations. Big data technologies allow us to take informed decisions from structured or unstructured data. Management and analysis of heterogeneous data generated by various sources brings numerous challenges and diversity in solutions. The aim of this chapter is to discuss different opportunities, issues, and challenges of big data with the main focus on the Hadoop platforms. We provide a detailed survey of opportunities, challenges, and issues of Hadoop-based big data developments in terms of data locality, load balancing, heterogeneity issues, scheduling issues, in-memory computation, multiple query optimizations, and I/O issues. Taxonomy of these challenges and opportunities is also presented.

Software quality is the degree to which the software conforms to its requirements. The complexity of software is on the rise with the developments of smart cities due to the complex nature of these applications and environments. Big data and Internet of Things (IoT) are driving radical changes in the software systems landscape. Together, big data, IoT, smart cities, and other emerging complex applications have exacerbated the challenges of maintaining software quality. The big data produced by IoT and other sources is used in designing or operating various software machines and systems. One of the challenges of big data is data veracity, which could lead to inaccurate or faulty system behavior. The aim of this paper is to review the technologies related to software quality in the era of big data, IoT, and smart cities. We elaborate on software quality processes, software testing and debugging. Model checking is discussed with some directions on the role it could play in the big data era and the benefits it could gain from big data. The role of big data in software quality is explored. Conclusion is drawn to suggest future directions.

The use of open source software has increased tremendously in the last few decades paving the way for many innovations such as Internet of Things (IoT) and smart cities. The open data licenses have also become prevalent with the emergence of big data and relevant technologies. These developments have given rise to the “Share more—Develop less” culture, which in turn have raised new legal issues. The community has been developing many new licenses to address these emerging legal issues. However, selecting the right license is becoming increasingly difficult due to the licensing complexities and continuous arrival of new licenses. This chapter reviews notable open source and open data licenses and the suitability of these licenses for various kinds of data and software. Subsequently, we propose frameworks for the selection of open source software and open data licenses. Conclusions are drawn with recommendations for the future work.

The world has seen exponential data growth due to social media, mobility, E-commerce, and other factors. The issues related to avalanche of data being produced are immense and cover variety of challenges that need a careful consideration. The use of HPDA (High Performance Data Analytics) is increasing at brisk speed in many industries and has resulted in expansion of HPC market in many new territories. HPC (High Performance Computing) and big data are different systems, not only at the technical level, but also have different ecosystems. HPC systems are mainly developed for computationally intensive applications but recently data intensive applications are also among the major workload in HPC environment. Big data analytics have grown in different perspectives and have separate developer communities. As we head towards the exascale and smart infrastructure era, the necessary integration of big data and HPC is currently a hot topic of research but still at very infant stages. Both systems have different architecture and their integration brings many challenges. The aim of this work is to identify the driving forces, challenges, current and future trends associated with the integration of HPC and big data. This paper is an extension of our earlier work. We have reviewed programming models and frameworks of big data and HPC. The big data and HPC challenges in the exascale-computing era are discussed. Additional elaborations are provided on HPC and big data convergence research efforts and future directions are provided. The HPC-big data convergence architecture proposed in our earlier paper has been enhanced.

In this work, we propose a standard-based test execution platform for dynamically adaptable IoT networks in smart cities which affords a platform-independent test system for isolating and executing runtime tests. This platform uses the TTCN3 standard and considers both structural and behavioral adaptations. Moreover, our platform is equipped with a test isolation layer that reduces the risk of interference between testing processes and business processes. We also compute a minimal subset of test cases to run and efficiently distribute them among the execution nodes. The minimal subset of test cases is obtained using a smart generation algorithm which keeps old tests cases which are still valid and replaces invalid ones by new generated or updated test cases.

Internet of Things (IoT) is revolutionizing all spheres of our lives leading the way for us to evolve into smarter societies. Wireless sensor networks (WSNs) are an integral part of the IoT ecosystems. Reliability, resilience, and energy conservation are the three most critical WSN requirements. Fault tolerance ensures the reliability and the resilience of WSNs in case of failures. This paper proposes a hierarchical clustered dynamic source routing (HCDSR) technique to improve fault tolerance and energy-efficient routing for WSNs. A survey of fault tolerant and energy-efficient routing techniques for WSNs is given. A taxonomy of fault tolerant techniques is introduced. The proposed HCDSR is simulated and compared with LEACH (low energy adaptive clustering hierarchy) and DFTR (dynamic fault tolerant routing) protocols to evaluate its performance. The results show that HCDSR outperforms LEACH and DFTR in terms of the total network energy, the number of nodes alive after a given time, and the network throughput. Directions for future work are given.

This is a work in progress in which we are interested in testing security aspects of Internet of Things for smart cities. For this purpose we follow a model-based approach which consists in: modeling the system under investigation with an appropriate formalism; deriving test suites from the obtained model; applying some coverage criteria to select suitable tests; executing the obtained tests; and finally collecting verdicts and analyzing them in order to detect errors and repair them. The adopted formalism is based on the model of extended timed automata with inputs and outputs. We propose a conformance testing relation, the so-called extended timed input–output conformance relation—etioco. For test execution, we introduce a cloud-based architecture.