Skip to content

Latest commit

 

History

History
20 lines (10 loc) · 443 Bytes

README.md

File metadata and controls

20 lines (10 loc) · 443 Bytes

spark-data-pipeline

Elasticsearch Setup

i) Download the Elasticsearch 6.3.0 or latest version and unzip it.

ii) Run the following command.

    $ bin/elasticsearch

Getting Started:

Clone and run in local mode:

    $ git clone [email protected]:techmonad/spark-data-pipeline.git
    $ cd spark-data-pipeline
    $ sbt run