Episode 398: Apache Kudu with Adar Leiber Dembo

Hadoop Integration

Adar Leiber-Dembo discusses the integration of Apache Kudu within the Hadoop ecosystem, highlighting its compatibility with various components like HDFS and Spark. Kudu is designed to excel in both scans and point lookups, making it versatile for different workloads. Adar explains how Kudu integrates with SQL engines such as Impala and Apache Hive, enhancing its utility in data analytics.

We do try to be really good citizens and integrate with a lot of different systems.

--- Adar Leiber-Dembo

This adaptability allows Kudu to fit seamlessly into existing big data workflows, offering a robust solution for analytics 1 2.

   

Integration Tools

The conversation shifts to the specific tools and services that facilitate Kudu's integration for data processing and analytics. Adar Leiber-Dembo mentions the use of Apache Spark, which provides a high-level framework for data transformations, and Spark SQL for SQL-based queries. These tools, along with others like Presto and Apache Drill, are integral to Kudu's functionality.

Spark offers a much richer and higher level framework for doing data transformations.

--- Adar Leiber-Dembo

This integration with various SQL engines allows Kudu to support a wide range of analytical tasks, making it a valuable asset in the big data ecosystem 1 2.