Java TPC-C

This project is a Java implementation of the TPC-C benchmark.

========= Compiling

Use this command to compile the code and produce a fat jar.

mvn package assembly:single

======== Database

To create the tpcc schema in MySQL:

cd database
mysql -u root
> create database tpcc;
> use tpcc;
> source create_tables.sql
> source add_fkey_idx.sql

It is possible to load data without the foreign keys and indexes in place and then add those after loading data to improve loading times.

================================= Generating and loading TPC-C data

Data can be loaded directly into a MySQL instance and can also be generated to CSV files that can be loaded into MySQL later using LOAD DATA INFILE.

In tpcc.properties set the MODE to either CSV or JDBC.

To run the load process:

java -classpath target/tpcc-1.0.0-SNAPSHOT-jar-with-dependencies.jar com.codefutures.tpcc.TpccLoad

It is possible to load data into shards where the warehouse ID is used as a shard key. The SHARDCOUNT and SHARDID properties must be set correctly when generating or loading data.

This option requires the use of a JDBC driver that supports automatic sharding, such as dbShards (http://www.dbshards.com).

=========================== Running the TPC-C Benchmark

Review the TPC-C settings in tpcc.properties, then run this command To run the tpcc benchmarks:

java -classpath target/tpcc-1.0.0-SNAPSHOT-jar-with-dependencies.jar com.codefutures.tpcc.Tpcc

Bugs can be reported to [email protected].

(c) 2014 CodeFutures Corporation.