Skip to main content Accessibility help
×
  1. Home
  2. Browse subjects
  3. Computer Science
  4. Distributed, Networked and Mobile Computing
  5. Content listing

Refine search

Refine search

Access:
Content type:
Publication date:
Apply   From and To filters
Subject: Show more
Journals Show more
Publishers: Show more
Societies Show more
Series: Show more
Collections: Show more

Actions for selected content:

Select all | Deselect all
  • View selected items
  • Export citations
  • Download PDF (zip)
  • Save to Kindle
  • Save to Dropbox
  • Save to Google Drive

Save Search

You can save your searches here and later view and run them again in "My saved searches".

Please provide a title, maximum of 40 characters.
Cancel
×

2000 results in Distributed, Networked and Mobile Computing

Page 19 of 100

  • 1 - Introduction to Smart Grid

  • from Part I - Introduction
  • Sudip Misra, Samaresh Bera
  • Book:
    Smart Grid Technology
    Published online:
    20 October 2018
    Print publication:
    28 March 2018, pp 3-17

  • Summary

    A smart grid is conceptualized as the existing power grid supported by bi-directional communication networks. Therefore, with the help of communication networks, service providers have real-time information about energy supply and demand. Moreover, customers also have real-time price information, based on which they can consume energy in an optimized manner, so that the total energy consumption cost is reduced. In addition to communication networks, several distributed energy generation units (such as solar, wind, and combined heat power) are also integrated to deploy distributed energy management policies. These distributed units are known as micro-grids. Typically, a smart grid consists of the following components – micro-grid, smart meter, renewable energy sources, and plug-in hybrid electric vehicles (PHEVs) [1]. Figure 1.1 depicts a schematic view of the smart grid architecture. Table 1.1 presents the basic differences between the traditional power grid and the smart grid.

    Smart Grid Framework and Communication Model

    Smart grid framework was initially conceptualized by the National Institute of Standards and Technology (NIST) in 2009. Figure 1.2 shows the smart grid framework conceptualized by NIST in its framework version 1.0 (adopted from [2]). The smart grid framework includes several entities – operators, markets, customers, utility providers, generation units, transmission units, and distribution units. The role of operators includes monitoring of the overall smart grid system and implementation of policies, while considering the interests of other entities. On the other hand, utility providers act as middle-persons between the customers and the energy markets. Therefore, it is possible to

    Distribution side includes local substations and customers. Further, the local substations are built with modern technologies to integrate energy from the main power grid with energy generated locally at the substation in a distributed manner. Therefore, the power supply to the customers is done in a semidistributed manner. The local substation reports to the main power grid about the local energy generation and demand from the customers.

    Customers’ homes are equipped with ‘smart meters,’ which send the energy consumption report to the substations. Further, the appliances installed at the customers’ ends are connected to the smart meters, in order to have smart energy monitoring system.

  • Foreword

  • Sudip Misra, Samaresh Bera
  • Book:
    Smart Grid Technology
    Published online:
    20 October 2018
    Print publication:
    28 March 2018, pp xxi-xxiv

  • Summary

    The growing concern about climate change and energy conservation necessitates the existing power grid to be modernized. Consequently, bi-directional communication network is placed atop the existing power grid, named as smart grid, with an aim to reduce energy loss and to provide improved energy management. This makes smart grid a cyber-physical system with a strong intertwining between the software, communication, control, and physical components. The cyber-physical property of smart grid necessitates adequate energy and information management, while considering the associated security concerns. Further, millions of smart meters are expected to be deployed to collect real-time status of energy consumption, and the collected information should be processed in real-time. Consequently, smart grid requires support of advanced computing platform and data management tools to manage real-time energy requirements and to process real-time information, respectively. Cloud computing and big data technologies are the topics of interest among the scientific community due to their inherent features of advanced computation and data management.

    Unlike most smart grid books, Sudip Misra and Samaresh Bera in this book focus on the specific challenges that are presented in the smart grid while integrating multiple entities together, which can be potentially solved using cloud computing technology and advanced data management approaches. It covers main aspects of smart grid, touching energy management, information management, and security, by leveraging the advantages of cloud computing technology and advanced data management approaches. The book discusses the important issues, such as demand response, dynamic pricing, load balancing, and smart meter data management, and includes the treatment to deal with such issues. In brief, the attractive features of this book are catalogued below:

  • a) Fundamental concepts of smart grid, cloud computing, and big data analytics are discussed. Additionally, the commonly used mathematical models and tools in this domain are also given concisely.

  • b) Cloud computing technology for smart grid presents a comprehensive discussion on the benefits of cloud computing technology in the smart grid for improved energy management and security.

  • c) Data management approaches mainly focus on the information management in the smart grid.

  • d) Smart grid simulation tools and initiatives worldwide highlight the recent efforts and on-going projects to realize the smart grid technology in a large-scale.

  • 12 - PHEVs: Internet of Vehicles

  • from Part III - Smart Grid Data Management and Applications
  • Sudip Misra, Samaresh Bera
  • Book:
    Smart Grid Technology
    Published online:
    20 October 2018
    Print publication:
    28 March 2018, pp 189-206

  • Summary

    Plug-in hybrid electric vehicles (PHEVs) are attracting the interest of researchers due to their unique features such as real-time demand side management and green technology. PHEVs use a combination of electric energy and traditional fossil fuel. They have an electric motor and an internal combustion engine. It is expected that PHEVs will produce less amount of air pollution compared to traditional vehicles. Moreover, PHEVs can charge and discharge their batteries, depending on the real-time energy status of the grid, which is widely termed as vehicle-to-grid (V2G) and grid-to-vehicle (G2V) transfer.

    Vehicle-to-grid (V2G) transfer is defined as the energy transfer from an electric vehicle to the power grid. In smart grid, the energy supply from generators and demand from its customers are dynamic over time. Therefore, sometimes, the grid may face supply–demand imbalance, which, in turn, may lead to grid failure. Using the concept of V2G transfer, energy can be supplied to the grid from the vehicle when there is a deficit in energy supply to the grid compared to the energy demand from its customers. Therefore, sudden energy demand peaks can be avoided using such technology.

    Vehicles also consume energy to charge their batteries. Consequently, they can charge their batteries when there is a surplus in energy supply to the grid. This energy transfer technology is termed as grid-to-vehicle (G2V) transfer. Therefore, surplus energy can be stored in the PHEVs’ batteries without wasting it. Hence, V2G and G2V techniques play an important role in real-time energy management in the smart grid. It is noteworthy that PHEVs are also integrated with different communication technologies, in order to have real-time energy supply–demand information. Consequently, suitable communication technology is also required for efficient operation of the PHEVs, while integrating them with the smart grid.

    Convergence of PHEVs and Internet of Vehicles

    Internet-of-vehicles (IoVs) are expected to play a major role in real-time information sharing and gathering information about roads and traffic. It is also useful to process, compute, and release the gathered information to other platforms. Based on the received data, the system can guide its users to take appropriate decisions. Moreover, it can provide Internet connectivity to its users as is done in vehicular ad-hoc networks (VANETs).

    Today's PHEVs will become tomorrow's IoVs with the energy management facilities. This is because, in addition to energy management, PHEVs can fulfill the objectives of the concept of IoVs.

  • Preface

  • Sudip Misra, Samaresh Bera
  • Book:
    Smart Grid Technology
    Published online:
    20 October 2018
    Print publication:
    28 March 2018, pp xxv-xxviii

  • Summary

    Overview and Goals

    Due to the growing concerns about climate change and energy conservation, there has been a greater focus than ever before on the development of digitalized and green technology for modern power delivery systems – Smart Grid – to the customers. Smart grid is conceptualized as the integration of bi-directional communication networks on the existing electricity networks. With the help of this type of communication networks, service providers and customers have real-time information about the energy consumption and price of energy, respectively. Therefore, service providers can maintain a balance between real-time supply and demand, so as to maximize their revenues. On the other hand, customers can consume energy in an optimized manner to minimize their energy consumption cost. Consequently, different intelligent mechanisms – demand response, dynamic pricing, integration of renewable energy sources, and distributed energy service – are introduced to support the smart grid architecture.

    Concurrently, cloud computing is an emerging technology that facilitates on-demand, real-time service platform dynamically with shared services to its clients. Cloud computing primarily provides three different services: software as a service (SaaS), infrastructure as a service (IaaS), and platform as a service (PaaS). Therefore, clients can utilize cloud resources according to their individual requirements. Another important feature of cloud computing is that it supports the implementation of a distributed architecture.

    Recently, different studies have revealed that cloud computing is a promising technology that can support the smart grid requirements from different perspectives: energy management, information management, and security services. Real-time energy management is the most important feature of the smart grid. Different schemes are introduced for this purpose: demand scheduling, integration of storage devices, cooperative architecture, are a few examples. A micro-grid in smart grid provides electricity to customers in a distributed manner with the help of renewable and non-renewable energy sources. Therefore, all entities (such as customers, service providers, and third parties) require adequate real-time information in order to execute the said schemes. Consequently, it is required to have a common platform from where all the entities can fetch adequate real-time information, which, in turn, necessitates the availability of suitable information management schemes. In the smart grid, a massive number of smart meters is expected to be deployed in order to have real-time information about customers’ energy consumption patterns. Presently, big data management approaches have been popular for processing and storage of massive volumes of heterogeneous data generated at high speeds.

  • 8 - Virtual Power Plant

  • from Part II - Cloud Computing Applications for Smart Grid
  • Sudip Misra, Samaresh Bera
  • Book:
    Smart Grid Technology
    Published online:
    20 October 2018
    Print publication:
    28 March 2018, pp 106-117

  • Summary

    In a smart grid, distributed energy generators are expected to play an important role in providing cost-effective and green energy-services to customers. Implementation of wind farms and solar panels are two popular distributed energy generating sources. In addition to this, combined heat power (CHP) and fuel cells are also expected to take part in the distributed energy generation process. However, due to the intermittent behavior of such distributed generators, maintaining reliable energy-services to customers is a big challenge for service providers. Additionally, energy generation from the distributed energy sources are environment dependent. Therefore, it may not be cost-effective if more energy is generated during off-peak periods. Consequently, it is required to have some kind of storage devices that can store energy on a situation basis (i.e., during off-peak hours), and can discharge energy to the distribution side during on-peak hours. In order to meet these requirements, a dynamic energy consumption/distribution platform, called virtual power plant (VPP), is considered.

    Concept of Virtual Power Plant

    A virtual power plant (VPP) is conceptualized as a combination of different distributed energy resources (DERs). Therefore, VPP can be considered a decentralized energy resource system with a large number of small-scale DERs such as solar energy, wind energy, CHPs, fuel cells, and plug-in hybrid electric vehicles (PHEVs). Consequently, it creates a single energy monitoring profile, while considering different parameters characterizing each distributed energy source. Figure 8.1 shows a schematic diagram of a virtual power plant consisting of different energy resources. As a whole, a VPP is capable of providing reliable and cost-effective energy supply to customers. Additionally, a large-scale VPP can be formed with a combination of different small-scale VPPs, as shown in Figure 8.2. Depending on various characteristics, there are two types of VPP – commercial and technical – that will be discussed in this section.

    Commercial VPP

    The concept of a commercial virtual power plant (CVPP) is used for energy trading in energy markets. With the integration of CVPP, energy-market risks can be reduced in order to utilize the distributed energy resources. Consequently, individual DER can benefit from an intelligent market in order to maximize its own revenue. A conceptual view of the commercial virtual power plant (CVPP) is shown in Figure 8.3. The operation of a CVPP consists of three phases – input, CVPP operation, and output. Depending on different inputs, CVPP optimizes revenue from individual DERs and offers different services.

  • 6 - Geographical Load-Balancing

  • from Part II - Cloud Computing Applications for Smart Grid
  • Sudip Misra, Samaresh Bera
  • Book:
    Smart Grid Technology
    Published online:
    20 October 2018
    Print publication:
    28 March 2018, pp 81-95

  • Summary

    The distributed architecture of smart grid facilitates micro-grids to provide electricity to the end-users in a distributed manner. Additionally, the micro-grids provide electricity to the customers using a combination of renewable and non-renewable energy sources. Therefore, each micro-grid has distributed generation facilities (such as combined heat power, solar, and wind), which act as the renewable and non-renewable energy sources. Consequently, micro-grids maintain a balance between real-time energy supply and demand to them. In order to have real-time information of the supply and demand, data centers need to be deployed for each micro-grid with bi-directional communication facility. In such a scenario, distributed load-balancing among the energy sources and data servers is required. Load-balancing is a technique by which massive load on a particular micro-grid can be reduced significantly.

    Need for Load–Balancing in Smart Grid

    There is a need to implement load-balancing in a smart grid due to the following reasons.

  • • As micro-grids provide electricity to customers in a distributed manner, the energy management policy implementation also takes place in a distributed manner at the micro-grids’ end. Therefore, it may happen that in a particular micro-grid's area, demand from the customers is very high. Consequently, real-time energy load on that micro-grid is very high, and it has to provide adequate electricity to the customers in order to meet the energy requirements of the latter. As a result, the micro-grid has to buy electricity from the main grid (through a market pricing policy) in order to meet the requirements. However, at the same time, it may happen that real-time demand is very low in another micro-grid. Therefore, there is an energy excess in that micro-grid. Consequently, it sells the surplus energy to the main grid. In such an energy exchange mechanism, a massive amount of energy is lost due to transmission loss. However, with the implementation of the load-balancing scheme, service areas of the micro-grids can be dynamically changed in order to maintain the balance between real-time energy supply and demand.

  • • Similar to the energy scarcity problem in a micro-grid's area, real-time information propagation may lead to the creation of a massive data load on the servers. With the help of bi-directional communication in the smart grid, customers communicate with the service providers for energy consumption. Therefore, real-time information propagation at the data centers also increases due to the massive energy demand from the customers.

  • 7 - Dynamic Pricing

  • from Part II - Cloud Computing Applications for Smart Grid
  • Sudip Misra, Samaresh Bera
  • Book:
    Smart Grid Technology
    Published online:
    20 October 2018
    Print publication:
    28 March 2018, pp 96-105

  • Summary

    With the implementation of bi-directional communication facility, service providers have real-time information about energy consumption by the customers and energy production by the generators. In such cases, dynamic pricing based mechanisms are more appealing than traditional static pricing based ones. Service providers decide real-time price based on the demand from customers or supply from the generation side or both the supply and demand. Consequently, customers have the opportunity to use electricity according to their preferred real-time price decided by the grid. Additionally, service providers maintain the real-time supply–demand curve by controlling the real-time price. When demand increases, the price may be increased to reduce the extra load from the grid, and vice-versa. Customers also schedule their energy consumption depending on the real-time price decided by the grid. Therefore, dynamic pricing policy is one of the important features of the smart grid that provides dynamic load balancing in the smart grid.

    Deployment of Dynamic Pricing in Smart Grids

    As the name suggests, in dynamic pricing, the price of energy varies over time according to the energy supply or demand or both. To implement the dynamic energy pricing policy, we need to address several challenges that will be discussed in this section.

    Determination of actual timeslot

    We discussed earlier that there are three different time-slots in the smart grid – on-peak, mid-peak, and off-peak. Based on the past days’ behavior, state (on, mid, or off-peak) of a particular time-slot is determined. For example, the office hour can be considered as on-peak, the morning and evening can be considered as mid-peak, and night can be considered as off-peak. However, in real-time, the situation in the grid may be different. For example, energy demand on a particular day may be different from another day. Moreover, as the smart grid is equipped with renewable energy sources, which are intermittent in nature, energy supply to the grid may also vary over multiple days. Therefore, it is not possible to blindly say that a particular time period is mid-peak, or on-peak, or off-peak. Consequently, it is important to define suitable strategies to determine the state of a time-slot.

  • 9 - Advanced Metering Infrastructure

  • from Part II - Cloud Computing Applications for Smart Grid
  • Sudip Misra, Samaresh Bera
  • Book:
    Smart Grid Technology
    Published online:
    20 October 2018
    Print publication:
    28 March 2018, pp 118-146

  • Summary

    Advanced metering infrastructure (AMI) is an important component of a smart grid that can help in fulfilling the objectives of the latter. It is a combination of smart meters and bi-directional communication networks. With the help of AMI, customers, service providers, and other entities exchange information among themselves in order to have a well-balanced smart grid environment. Consequently, with the wide deployment of smart meters supported by bi-directional communication networks, it is easier to monitor real-time energy supply–demand information, automatic billing, and many others. Moreover, it is one of the primary requirements for establishing a smart grid environment.

    Requirements

    Typically, an AMI system should meet the following requirements [1]:

  • • Data storage: Energy consumption data is required to be stored in the smart meter. Therefore, the AMI should help smart meters to store real-time and past energy consumption data. To fulfill this requirement, adequate memory should be installed in the smart meters. Moreover, it is also required to ensure that the stored data is accessible by only authenticated users.

  • • Communication: Different communication protocols are also required to communicate with different entities; examples include the IEEE 802.15.4, IEEE 802.11, and wide area network protocols. For example, home appliances may communicate with the home energy management unit (HEMU) using the IEEE 802.15.4 protocol. On the other hand, a HEMU may communicate with the local aggregator unit using the IEEE 802.11 protocol. Finally, the aggregator units may communicate with the service provider through the protocols used in wide area networks. Therefore, suitable communication protocols are required to be deployed, taking into consideration both their advantages and disadvantages.

  • • Management: Different management functions include remote management, billing management, and load control. All these tasks should be performed in an adaptive manner without the involvement of human beings. In case of remote management, fault in the smart meter must be rectified, and the corresponding action should be taken in order to avoid service disruption. In case of billing management, customers can be automatically billed according to their energy consumption. Finally, in case of load control, energy demand from some of the smart meters can be discarded to establish a balance between the energy supply and demand.

  • • User interface: It can be used as an alternative to users for accessing AMI information; the interface includes LCD displays, graphics constructions, keyboard, and a website with authorization facility.

  • 13 - Smart Buildings

  • from Part III - Smart Grid Data Management and Applications
  • Sudip Misra, Samaresh Bera
  • Book:
    Smart Grid Technology
    Published online:
    20 October 2018
    Print publication:
    28 March 2018, pp 207-224

  • Summary

    Buildings are one of the most prominent energy consumers in modern-day power grids. In particular, commercial buildings consume large amounts of energy for operating lighting systems, heating–ventilation and air-conditioning (HVAC), and IT equipment including servers. High energy consumption at the buildings increases energy consumption cost and environment pollution. To deal with these issues, buildings can be converted to ‘smart buildings’. Through the concept of smart buildings, electric loads can be controlled in an adaptive manner. Unused equipment can be switched off so that the total energy consumption in the building can be reduced. The appliances that are installed inside a building will generate huge data in order to have real-time information inside the building. To fulfill these objectives, different methodologies exist. In this chapter, we focus on commercial buildings as smart buildings for efficient energy management.

    Concept of Smart Building

    Most commercial buildings are equipped with HVAC systems, IT equipment, servers, and plug-in equipment. These equipments can be integrated with different sensing and actuation technologies, so that they can be controlled dynamically depending on their requirements. Different useful techniques that are deployed in different buildings to make them smart buildings, are discussed here.

  • • Sensing: It is the key technology required for setting up a smart building. Different sensors are installed in the building (inside and outside), so that different parameters (such as temperature, smoke, humidity, and motion) can be sensed in real-time. The sensed data are forwarded to the data center network for processing, computing, and decision making. For data collection and forwarding, wireless sensor network (WSN) is an emerging technology that can be used due to its unique features such as the capability of operating in low power. Typically, IEEE 802.15.4-based technologies (such as Zig-Bee and 6LowPAN) are used in WSN to forward real-time data. User context, environment, and energy consumption are three major factors to be monitored to establish a smart building. The sensors that are installed to monitor these components generate huge amount of data. This also requires suitable data management techniques for real-time energy management inside the building, as discussed in Chapter 11.

  • • Actuation: After sensing the users’ contexts, environment, and energy consumption, actuation is another important task to be employed. HVACs should be controlled in an efficient manner considering the sensed information received from individual sensors.

  • 5 - Demand Response

  • from Part II - Cloud Computing Applications for Smart Grid
  • Sudip Misra, Samaresh Bera
  • Book:
    Smart Grid Technology
    Published online:
    20 October 2018
    Print publication:
    28 March 2018, pp 61-80

  • Summary

    Typically, in traditional power delivery systems, energy is supplied to the customers using non-renewable energy sources. Therefore, the energy generators are based on fossil fuels. Consequently, carbon emission in the environment is very high, and there is potentially a great impact on the environment. In contrast, in the smart grid system, energy is supplied to the end-users using renewable and non-renewable energy sources. Additionally, the implementation of distributed energy sources plays an important role for real-time energy management in the smart grid. Consequently, in the presence of distributed energy sources at the customers’ end, electricity consumers can schedule their shiftable appliances to the off-peak hours to relieve the extra load on the grid during on-peak hours. This mechanism is widely known as demand response (DR). The objective of any demand response mechanism is to reduce energy consumption cost to the customers and reduce imbalance between the energy supply and demand, while modifying the energy consumption pattern according to real-time price signals.

    Fundamentals of Demand Response and Challenges

    Different technologies are introduced in a smart grid system in order to establish the demand response mechanism – demand scheduling, online energy monitoring, and deployment of distributed energy sources. As mentioned earlier, customers can schedule their shiftable appliances to off-peak hours so as to minimize their energy consumption cost. Additionally, the real-time energy demand to the grid can also be relieved during peak hours. In the traditional power delivery system, such strategies cannot be deployed by the service providers in order to schedule the customers’ energy demand. Therefore, we have on-peak and off-peak hours in which energy demand is more and less, respectively, than the energy supply to the grid. On the other hand, using the demand response mechanism, the energy demand from the customers can be modified according to the energy supply to the grid in order to minimize the imbalance between the energy supply and the demand to the grid. However, there are several challenges that should be considered in order to install demand response mechanisms in real-life environments.

  • 1. Peak load reduction: In the smart grid, there are three different time-slots depending on the energy load on the grid – on-peak, mid-peak, and off-peak. During on-peak hours, energy demand from the customers is high. On the other hand, during mid-peak and off-peak hours, demands from the customers are moderate and low, respectively.

  • 10 - Cloud-Based Security and Privacy

  • from Part II - Cloud Computing Applications for Smart Grid
  • Sudip Misra, Samaresh Bera
  • Book:
    Smart Grid Technology
    Published online:
    20 October 2018
    Print publication:
    28 March 2018, pp 147-166

  • Summary

    Smart grid is essentially a cyber-physical system (CPS) integrated with electric distribution supported with bi-directional communication networks. To support the communication and computing requirements, third party service providers are also expected to participate. Consequently, the third party service providers also access customers’ information that is sent through smart meters and backbone networks. As a result, the smart grid system should have appropriate security and privacy policies to secure the smart meters data and other components from unauthorized access. It is a challenging issue to secure the smart grid from cyber attacks in the presence of online connectivity of all components with the Internet [1]. Among different threats that are present in the smart grid in terms of security, energy theft is one of the most important issues to consider. The intruder can access the control point of the smart meters, and thereby, modify real-time information to change energy consumption information reported to utility providers. Therefore, adequate security policies are needed to deal with such issues. In this chapter, we will discuss general security requirements in a communication network. Then specific security issues present in the smart grid and their remedies are discussed.

    Security in Data Communication

    In data communication, at least two parties (the sender and the receiver) are involved in message exchange. The sender sends a message for an intended receiver. On receiving the message, the receiver takes adequate actions or decisions. In such a scenario, if the sender sends the message directly to the receiver, it may happen that the message is modified or accessed by an attacker. However, the receiver assumes that the received message is the original one sent by the sender. Additionally, it may also happen that an attacker uses the sender's identity to send malicious data to the receiver. Finally, after sending a message, the sender should not be able to repudiate the sent message. Therefore, it is required to implement security mechanisms to prevent such incidents. For example, in banking transactions, appropriate security mechanisms should be in place so that no one other than the actual user can withdraw money. In this chapter, we focus on the cryptography-based security aspects in data communication instead of focusing on other aspects such as intrusion detection and denial-of-service (DoS) attacks. The basic terms used in security are as follows.

  • 3 - Introduction to Big Data Analytics

  • from Part I - Introduction
  • Sudip Misra, Samaresh Bera
  • Book:
    Smart Grid Technology
    Published online:
    20 October 2018
    Print publication:
    28 March 2018, pp 38-48

  • Summary

    Big data is conceptualized as a large data set with high complexity that cannot be managed with the traditional data processing schemes in an adequate manner. Big data technology is described as a holistic information management of new types of data alongside traditional data [1]. For example, traditionally, the backbone network carries different homogeneous information generated by end-users. In addition to this, due to the deployment of more number of sensors, actuators, and other different data sources, the network needs to carry heterogeneous information. It is also expected that the same network channel would be used to carry this heterogeneous information. The challenges in big data analytics include analyzing, capturing, processing, searching, storing, and visualizing the large data set. The term ‘big data’ refers to predictive analysis, i.e., predicting the future condition based on the results obtained from processing existing data. Thus, we can predict the system status in advance to take adequate measures to deal with situations in real-time.

    According to the existing literature, there is no hard and fast definition of big data. As mentioned earlier, big data is a conceptualized term that is used to represent a large data set, which is complex and heterogeneous in nature. Moreover, it is hard to process such large data set using the traditional applications. Due to the growing need to digitize data, it is required to store and process large amounts of data generated from heterogeneous devices. To get an idea about the amount of data generated over several past years, Figure 3.1 presents certain statistics regarding growth of information storage capacity with digitization of data (adopted from [2]). As can be seen in Figure 3.1, there is a large amount of data being stored in digital storage devices. However, we need efficient processing applications to process such large amount of data before storing it for future use. Different attributes of big data are discussed in the subsequent sections.

    Attributes of Big Data

    As the name suggests, one of the main attributes of ‘big data’ is volume, i.e., the size of the data. However, it is not only about the volume, but also other factors such as velocity and variety of the data being analyzed. Figure 3.2 presents the attributes diagrammatically. Let us discuss the attributes in detail.

Page 19 of 100