Data Science and Big Data Analytics in Smart Environments

Marta Chinnici, Florin Pop, Catalin Negru, Marta Chinnici, Florin Pop, Catalin Negru

• Publisher: CRC Press
• Year: 2021
• Print ISBN:
  9780367776039
  9780367407131
• Edition: 1
• eBook ISBN:
  9781000386059
• Topic: Computer Science
• Subtopic: Statistics for Business & Economics
• Index: Computer Science

Most applications generate large datasets, like social networking and social influence programs, smart cities applications, smart house environments, Cloud applications, public web sites, scientific experiments and simulations, data warehouse, monitoring platforms, and e-government services. Data grows rapidly, since applications produce continuously increasing volumes of both unstructured and structured data. Large-scale interconnected systems aim to aggregate and efficiently exploit the power of widely distributed resources. In this context, major solutions for scalability, mobility, reliability, fault tolerance and security are required to achieve high performance and to create a smart environment. The impact on data processing, transfer and storage is the need to re-evaluate the approaches and solutions to better answer the user needs. A variety of solutions for specific applications and platforms exist so a thorough and systematic analysis of existing solutions for data science, data analytics, methods and algorithms used in Big Data processing and storage environments is significant in designing and implementing a smart environment.

Fundamental issues pertaining to smart environments (smart cities, ambient assisted leaving, smart houses, green houses, cyber physical systems, etc.) are reviewed. Most of the current efforts still do not adequately address the heterogeneity of different distributed systems, the interoperability between them, and the systems resilience. This book will primarily encompass practical approaches that promote research in all aspects of data processing, data analytics, data processing in different type of systems: Cluster Computing, Grid Computing, Peer-to-Peer, Cloud/Edge/Fog Computing, all involving elements of heterogeneity, having a large variety of tools and software to manage them. The main role of resource management techniques in this domain is to create the suitable frameworks for development of applications and deployment in smart environments, with respect to high performance. The book focuses on topics covering algorithms, architectures, management models, high performance computing techniques and large-scale distributed systems.

Chapter 1
Mobility-Aware Solutions for Edge Data Center Deployment in Urban Environments

Piergiorgio Vitello,^a,* Andrea Capponi,^a Claudio Fiandrino^b and Guido Cantelmo^c

^a University of Luxembourg.

^b IMDEA Networks Institute.

^c Technical University of Munich.

^* Corresponding author: [email protected]

1.1 Introduction

The fifth generation (5G) of cellular mobile networks is tailored to support high mobility, massive connectivity, extremely high data rates and ultra low latency. For this, it relies on Software-Defined Networking (SDN) and Network Function Virtualization (NFV). Radio access and core functions are virtualized and executed in edge data centers (EDCs) according to the Multi-access Edge Computing (MEC) principle. MEC was formerly known as Mobile Edge Computing and is standardized by the European Telecommunications Standards Institute (ETSI) [137]. MEC provides computing service closer to the end user, and thus finds applicability in scenarios where locality and low-latency are essential¹ [343]. MEC is not tied to a single radio technology, but embraces both cellular and other radio access technologies such as WiFi and is agnostic to the evolution of the mobile network itself (4G or 5G).

1 In the rest of the chapter the words user, mobile user, and citizen are used interchangeably.

The edge, also known as MEC host, is a (nano) data center deployed within the mobile network operator (MNO) infrastructure. The edge provides computing functionalities, can aggregate virtualized core and radio network functions and enables resource-constrained mobile devices to prolong battery lifetime while enhancing and augmenting performance of the mobile applications [119]. To this date, the research on edge computing has mainly focused on resource management and allocation by trading power consumption and communication delays [94, 381] while seminal works have mainly focused on the definition of architectural design principles [308, 261]. Emulation platforms for research in the area have only started to appear recently [122] and little attention has been paid to the problem of EDC deployment.

The EDC deployment is a particularly interesting and challenging problem in the context of smart cities. In spite of this, only a vision paper has explored such areas by assessing the feasibility of leveraging three different infrastructures, i.e., cellular base stations, routers and street lamps and analyzing the potential city coverage if only a subset of these elements was upgraded to furnish EDC capabilities [132]. This study is a step forward to solve the problems of coverage, EDC selection and user-to-EDC assignment. However, it does not consider urban mobility and the energy-cost of operating such infrastructure (e.g., to power on servers, base stations). Citizens mobility is influenced by many factors, including trip purposes and geographically imposed restrictions, such as home and work locations. These complex phenomena (herein called urban dynamics) influence the computational demand of BSs and EDCs that changes over time. A recent study unveils a strong correlation between the urbanization tissue, specifically land use, and the average traffic volume per-user by highlighting that different mobile services typically represent different temporal behaviors but uniform spatial patterns [129].

This chapter tackles the problem of EDCs deployment in a smart city context leveraging two observations. First, users access networks only with cellular connectivity and assume that EDCs should only be deployed at current Base Stations (BSs) sites to re-use already deployed infrastructure (e.g., power supply, cabinets on roofs). Thus, this solution is capital-expenditure free for MNOs. Second, this study focuses on human mobility. Within a city, complex dynamics regulate the inter-dependency of land use and citizen movements [179], i.e., the spatial distribution of citizens and locations they visit that determine mobility patterns. Similar to [397], this study exploits crowd sensed data (namely, Google Popular Times) to infer and predict human mobility with the goal of determining estimates of computing demand and the optimal EDC deployment that minimizes outages. Third, this analysis targets energy-efficiency as one of the objectives of deployment. As it is well known from research in cloud computing [121] and radio access network [141], it is possible to reduce the energy expenditure in the two domains. For example, consolidating jobs/tasks into the minimum number of servers or active BSs in order to idle those with low workloads.

For the problem of EDCs deployment, this study proposes and compares three heuristics, namely distributed deployment algorithm (DDA), mobility-aware deployment algorithm (MDA) and Energy-aware MDA (EMDA) which enforce energy savings by powering off base stations with low utilization. The heuristics are evaluated with simulations that show how mobility-aware techniques outperform those that do not consider citizens’ mobility. Simulators and emulators for fog computing/MEC are available. However, on the one hand, emulators provide a very fine precision of network and computing mechanisms that prevent assessing performance at city-level scale. On the other hand, simulators model either computing, networking aspect or both well while little support for realistic user mobility is enforced. This chapter specifically tackles this problem and proposes CrowdEdgeSim, the first simulator that simultaneously models urban mobility, network and computing dynamics. CrowdEdgeSim is key to assessing performance of MEC components, e.g., resource management techniques, at city level scale for smart environments. This chapter exposes the architecture of Crowd-EdgeSim which features independent modules to model and simulate pedestrian mobility, mobile devices, network and computing resources and ETSI-compliant MEC applications. These three main factors influence each other in city-wide scenarios and are building blocks of the simulation environment, and require a further investigation on communications and MEC deployment. CrowdEdgeSim is based on the reference architecture of CrowdSenSim [120, 258], a s...

Table of contents

• Cover
• Title Page
• Copyright Page
• Preface
• Contents
• 1. Mobility-Aware Solutions for Edge Data Center Deployment in Urban Environments
• 2. Effective Data Assimilation with Machine Learning
• 3. Semantic Data Model for Energy Efficient Integration of Data Centres in Energy Grids
• 4. Managing the Safety in Smart Buildings using Semantically-Enriched BIM and Occupancy Data Approach
• 5. Belief Rule-Based Adaptive Particle Swarm Optimization
• 6. NoSQL Environments and Big Data Analytics for Time Series
• 7. A Territorial Intelligence-based Approach for Smart Emergency Planning
• 8. Big Data Analysis and Applications for Energy Performant Buildings and Smart Cities
• 9. Selecting Suitable Plants for a Given Area using Data Analysis Approaches
• 10. Ontology-Based Security Requirements Framework for Current and Future Vehicles
• 11. Dynamic Resource Provisioning using Cognitive Intelligent Networks based on Stochastic Markov Decision Process
• 12. Data Model for Water Resource Management
• References
• Index