Skip to main content

2019 | Buch

Real-Time & Stream Data Management

Push-Based Data in Research & Practice

insite
SUCHEN

Über dieses Buch

While traditional databases excel at complex queries over historical data, they are inherently pull-based and therefore ill-equipped to push new information to clients. Systems for data stream management and processing, on the other hand, are natively push­oriented and thus facilitate reactive behavior. However, they do not retain data indefinitely and are therefore not able to answer historical queries.
The book will first provide an overview over the different (push-based) mechanisms for data retrieval in each system class and the semantic differences between them. It will also provide a comprehensive overview over the current state of the art in real-time databases. It will first include an in-depth system survey of today's real-time databases: Firebase, Meteor, RethinkDB, Parse, Baqend, and others. Second, the high-level classification scheme illustrated above provides a gentle introduction into the system space of data management: Abstracting from the extreme system diversity in this field,

Inhaltsverzeichnis

Frontmatter
Chapter 1. An Introduction to Real-Time Data Management
Abstract
In recent years, users have come to expect reactivity from their applications, i.e. they assume that changes made by other users are immediately reflected in the interfaces they are using. Examples are shared worksheets and websites targeting social interaction. These applications require the underlying data storage to publish new and updated information as soon as it is created: Data access is push-based. In contrast, traditional database management has been tailored towards pull-based data access where information is only made available as a direct response to a client request.
Wolfram Wingerath, Norbert Ritter, Felix Gessert
Chapter 2. Database Management
Abstract
The first databases were hierarchical and network databases, developed during the 1960s. They exposed procedural query interfaces (as opposed to descriptive ones), so that accessing specific information in one of these systems was similar to navigating to a specific file within a file system.
Wolfram Wingerath, Norbert Ritter, Felix Gessert
Chapter 3. Real-Time Databases
Abstract
While traditional database systems are targeted at providing a consistent snapshot of the application domain, real-time databases acknowledge that data may evolve. Both the architectures and client APIs of real-time databases reflect that facts can change over time and that the system may have to enhance or correct issued information.
Wolfram Wingerath, Norbert Ritter, Felix Gessert
Chapter 4. Data Stream Management
Abstract
In some domains, data arrives so fast and in such great quantity that storing it in a database collection is simply infeasible. When the incoming data relates to ongoing (real-world) events that require immediate action, persistence may further not even be useful; for example, data in electronic trading, network monitoring, or real-time fraud detection is only valuable for a short amount of time and therefore has to be utilized immediately. To adapt to these circumstances, data stream management systems (DSMSs) introduce the data stream as an abstraction for an infinite sequence of database records that arrive over time. The raw data streams arriving at the systems are usually referred to as base streams, whereas those resulting from data transformations (e.g. queries) are called derived streams. Since a data stream is impossible to store entirely due to its unbounded nature, DSMSs drop the database requirement of eternal data persistence: They retain incoming records for limited time only and eventually discard them.
Wolfram Wingerath, Norbert Ritter, Felix Gessert
Chapter 5. General-Purpose Stream Processing
Abstract
Unlike data stream management systems that are mostly intended for analyzing structured information through declarative query languages, systems for stream processing expose generic and imperative (i.e. non-declarative) programming interfaces to work with structured, semi-structured, and entirely unstructured data. Rather than yet another approach for querying data, stream processing can thus be seen as the latency-oriented counterpart to batch processing. In this chapter, we provide an overview over some of the most popular distributed stream processing systems currently available and highlight similarities, differences, and trade-offs taken in their respective designs.
Wolfram Wingerath, Norbert Ritter, Felix Gessert
Chapter 6. State of the Art and Future Directions
Abstract
The ability to notify clients of changes to their critical data has become an important feature for both data storage systems and application development frameworks. In the final chapter of this book, we summarize the state of the art in push-based data access and identify possible next steps in the development of applications and technology in the field of real-time data management.
Wolfram Wingerath, Norbert Ritter, Felix Gessert
Metadaten
Titel
Real-Time & Stream Data Management
verfasst von
Wolfram Wingerath
Norbert Ritter
Felix Gessert
Copyright-Jahr
2019
Electronic ISBN
978-3-030-10555-6
Print ISBN
978-3-030-10554-9
DOI
https://doi.org/10.1007/978-3-030-10555-6

Neuer Inhalt