Kinesis Connector

Kinesis is Amazon’s fully managed cloud-based service for real-time processing of large, distributed data streams.

This connector allows the use of Kinesis streams as tables in Presto, such that each data-blob (message) in a kinesis stream is presented as a row in Presto. A flexible table mapping approach lets us treat fields of the messages as columns in the table.

Under the hood, a Kinesis shard iterator is used to retrieve the records, along with a series of getRecords calls. The shard iterator starts by default 24 hours before the current time and works its way forward (this is configurable now, so it can changed as needed). To be able to query a stream, table mappings are needed as explained in [Table Definition] (https://github.com/qubole/presto-kinesis/wiki/Table-Definitions). These table definitions can be stored on Amazon S3 (preferred) or stored in a local directory on each Presto node.

This version has been updated to work with Presto 0.149 and the latest version of the Amazon AWS SDKs. Other improvements and bug fixes have also been made.

This connector is Read-Only connector. It can only fetch data from kinesis streams, but can not create streams or push data into the al ready existing streams.

Building

mvn clean package

This will create target/presto-kinesis-<version>-bundle.tar.gz file which contains the connector code and its dependency jars.

Installation

You will need to augment your presto installation on coordinator and worker nodes to make sure the connector is loaded and configured properly. We will use $PRESTO_HOME to refer to the presto installation directory.

This distribution contains a sample kinesis.properties file that can be used as a starting point (in etc/catalog). Comments above each property summarize what each property is for.