Название: Distributed Systems: Theory and Applications
Автор: Ratan K. Ghosh, Hiranmay Ghosh
Издательство: Wiley-IEEE Press
Год: 2023
Страниц: 563
Язык: английский
Формат: pdf (true)
Размер: 10.1 MB

Distributed Systems Comprehensive textbook resource on distributed systems―integrates foundational topics with advanced topics of contemporary importance within the field.

Distributed Systems: Theory and Applications is organized around three layers of abstractions: networks, middleware tools, and application framework. It presents data consistency models suited for requirements of innovative distributed shared memory applications. The book also focuses on distributed processing of big data, representation of distributed knowledge and management of distributed intelligence via distributed agents. To aid in understanding how these concepts apply to real-world situations, the work presents a case study on building a P2P Integrated E-Learning system. Downloadable lecture slides are included to help professors and instructors convey key concepts to their students.

Conventionally in distributed computing, all compute-intensive processing that an end-user system cannot execute is relocated to peers offering the desired service. It relies heavily on the paradigm of client-server computing. With the availability of improved network technology and cloud computing, we can integrate many innovations in the development of distributed applications. Instead of using monolithic servers, distributed applications now make use of cloud services. With pay-per-use options, we can request an array of cloud-hosted services known as microservices. Typically, Internet of Things (IoT)-based automation applications are assembled as a collection of microservices relying on mobile agent technology on slow networks like 3G/4G. Microservice-based applications have a lot of dependencies and may make use of different programming artifacts. Microservices are accessed using REST Application Programming Interfaces (APIs) via Hyper Text Transfer Protocol (HTTP) transport. Typically, such applications are containerized or packaged as executable images to make them run on any host device.

Data processing systems have traditionally depended on a relational database to deal with structured data. However, a relational database is not suitable for storing unstructured data, such as text, images and sensor data. Further, relational databases use “join” operations extensively between multiple tables to create a required view of data. The various tables may be stored in different locations in a distributed storage system, making a join operation computationally expensive and slow. Some distributed relational database systems use shared-disk architecture, where all processing nodes access the same data from a central repository hosted on a distributed storage system. These systems provide consistent data at all times but do not scale well. Moreover, a relational database relies on fixed schema definitions that are difficult to extend when new data types are included. The limitations of relational databases have led to the development of other data models for distributed big data storage. They are collectively known as NoSQL databases. Some authors prefer to interpret the term “NoSQL” as “not only SQL,” not to preclude relational databases, which are also useful in big-data systems.

Additional topics discussed in Distributed Systems: Theory and Applications include:

Network issues and high-level communication tools
Software tools for implementations of distributed middleware.
Data sharing across distributed components through publish and subscribe-based message diffusion, gossip protocol, P2P architecture and distributed shared memory.
Consensus, distributed coordination, and advanced middleware for building large distributed applications
Distributed data and knowledge management
Autonomy in distributed systems, multi-agent architecture
Trust in distributed systems, distributed ledger, Blockchain and related technologies.

Researchers, industry professionals, and students in the fields of science, technology, and medicine will be able to use Distributed Systems: Theory and Applications as a comprehensive textbook resource for understanding distributed systems, the specifics behind the modern elements which relate to them, and their practical applications.