Hadoop-Spark-SQL

API Endpoints

• Base url is http://localhost:8080/cloudproject.
• GET request on /api/test : Returns a page displaying Hello World!. Useful for testing if the API is live
• POST request on /api/query : Request body should be in application/json format.
  • Input Parameter : query (String) -> the SQL query to be run
  • Output : application.json containing the required output parameters.

Configuration

Global configuration file: com.project.cloud.Globals

• This file contains HDFS output paths, which can be configured.

Project configuration: pom.xml

• Maven project file for build settings and dependencies.

Requirements

• The project, built into a webapp has been tested on tomcat8.5.
• It may be required to configure the amount of memory allocated to Tomcat JVM.

Assumptions

• <Condition2> of the WHERE clause in INNER JOIN query is assumed to be an equality operation in one of the columns of the final table.
• <COLUMNS> of SELECT and GROUP BY have been assumed to be the same.
• Input value of any Aggregate Function, and the value against which it is compared in HAVING clause is assumed to be integer.