Book Image

Data Lake for Enterprises

By : Vivek Mishra, Tomcy John, Pankaj Misra
Book Image

Data Lake for Enterprises

By: Vivek Mishra, Tomcy John, Pankaj Misra

Overview of this book

The term "Data Lake" has recently emerged as a prominent term in the big data industry. Data scientists can make use of it in deriving meaningful insights that can be used by businesses to redefine or transform the way they operate. Lambda architecture is also emerging as one of the very eminent patterns in the big data landscape, as it not only helps to derive useful information from historical data but also correlates real-time data to enable business to take critical decisions. This book tries to bring these two important aspects — data lake and lambda architecture—together. This book is divided into three main sections. The first introduces you to the concept of data lakes, the importance of data lakes in enterprises, and getting you up-to-speed with the Lambda architecture. The second section delves into the principal components of building a data lake using the Lambda architecture. It introduces you to popular big data technologies such as Apache Hadoop, Spark, Sqoop, Flume, and ElasticSearch. The third section is a highly practical demonstration of putting it all together, and shows you how an enterprise data lake can be implemented, along with several real-world use-cases. It also shows you how other peripheral components can be added to the lake to make it more efficient. By the end of this book, you will be able to choose the right big data technologies using the lambda architectural patterns to build your enterprise data lake.
Table of Contents (23 chapters)
Title Page
Credits
Foreword
About the Authors
About the Reviewers
www.PacktPub.com
Customer Feedback
Preface
Part 1 - Overview
Part 2 - Technical Building blocks of Data Lake
Part 3 - Bringing It All Together

Elasticsearch deployment options


Elasticsearch supports a scale-out deployment architecture, with the main priority on availability. The Elasticsearch cluster is composed of nodes playing master and data node roles. Quorum based availability drives cluster availability in general.

While the most common deployment is that of single data center deployment, which can comprise dedicated master nodes and dedicated data nodes, for larger clusters, multiple data center deployments are also required for very high availability of critical applications.

It is generally not a very good idea to deploy Elasticsearch clusters across data centers, since the Elasticsearch leader election algorithm and data node selection are based on network distance, which assumes that all the nodes are identical in terms of all resources; it is expected that in a cluster all nodes are equidistant. This can go wrong in cross data center clusters at times, and is not worth the risk.

Hence it is generally recommended to have...