Skip to content

Latest commit

 

History

History
47 lines (27 loc) · 1.33 KB

HOW-TO-RUN.md

File metadata and controls

47 lines (27 loc) · 1.33 KB

Instructions for running BenchmarkSQL

Depending on the DBMS, you need to modify a set of files to configure the connection properties and the environment.

There is a file in the root directory for each DBMS that describes the details to configure and execute BenchmarkSQL.

These are the global steps for any DBMS.

Requirements

  • JDK8 is required.
  • Driver to access the database.

Compile the BenchmarkSQL source code

This is a one-time step, to generate the jar file to execute BenchmarkSQL.

Create a database and the credentials to access

In this step the database is created in the DMBS and a set of credentials are generated.

Adjust the BenchmarkSQL configuration file

Modify the corresponding file under /run. Make sure you can connect to database with the credentials.

Build the schema and initial database load

Creates the tables and other objects in the database. Also, generates the data for each warehouse, or loads this data from files.

Run the configured benchmark

Run your first benchmark, to make sure everything is correct.

Scale the benchmark configuration

Tune your database for the workload.

Clean the environment

Once you have finished with the tests, you can clean the environment by deleting all data and objects created.