Description:
Spark is an analytics engine built on Hadoop that works with big data, data science and processing batch, and streaming data. In this 11-video course, discover how to develop applications in Spark to work with streaming data and explore different ways to process streams and generate output. Key concepts covered here include installing the latest version of PySpark; configuring a streaming data source using Netcat and writing applications to process the stream; and effects of using the Update mode for output of your stream processing application. Learn how to write an application to listen for new files added to a directory; compare the Append output to the Update mode and distinguish between the two; and develop applications that limit files processed in each trigger and use Spark's Complete mode for output. Next, learners perform aggregation operations on streaming data with the DataFrame API (application programming interface); work with Spark SQL to process streaming data by using SQL queries; and learn ways to use Spark for streaming data and ways to process streams and generate output.
Target Audience:
Duration: 00:53
Description:
Spark, an analytics engine built on Hadoop, can be used for working with big data, data science, and processing batch and streaming data. Explore the fundamentals of working with streams using Spark in this 9-video course. Key concepts covered here include the differences between batch and streaming data and the types of streaming data sources; processing streaming data, transformation of streams, and materialization of the results of the transformation; and how use of a message transport decouples a streaming application from the sources of streaming data. Next, learn about techniques used in Spark 1.x to work with streaming data and how it contrasts with processing batch data; how structured streaming in Spark 2.x is able to ease the task of stream processing for the app developer; and how streaming processing works in both Spark 1.x and 2.x. Finally, learn how triggers can be set up to periodically process streaming data and the various output modes available to publish the results of stream processing; and the key aspects of working with structured streaming in Spark.
Target Audience:
Duration: 00:51