-
Book Overview & Buying
-
Table Of Contents
Scalable Data Architecture with Java
By :
In this section, we will discuss two very famous patterns that support both batch and real-time processing. Since these patterns support both batch processing and stream processing, they are categorized as hybrid patterns. Let’s take a look at the most popular hybrid architectural patterns.
First, let’s understand the need for Lambda architecture. In distributed computing, the CAP theorem states that any distributed data can guarantee only two out of the three features of the data – that is, consistency, availability, and partition tolerance. However, Nathan Marz proposed a new pattern in 2011 that made it possible to have all three characteristics present in a distributed data store. This pattern is called the Lambda pattern. The Lambda architecture consists of three layers, as follows: