Unified Stack

Spark is a unified stack of components and libraries. Below graphic depicts where Spark fits in a BigData Application Stack.

On the left we see different data sources. The stack in the middle represents various Big Data processing workflows and tools that are commonly used. We can have just one of these workflows in your application, or a combination of many. Any of these workflows could read/write to or from the storage layer. As you can see here, with Spark, you have a unified stack. You can use Spark for any of the workflows. The output can then be used to create real-time dashboards and alerting systems for querying and advanced analytics; and loading into an enterprise data warehouse.