MySQL HeatWave TPCC

MySQL HeatWave is a fully managed and highly scalable in-memory database service which provides a cost-efficient solution for OLTP, OLAP and Machine Learning. It is available on both Oracle Cloud Infrastructure (OCI) and Amazon Web Service (AWS).

HeatWave is tightly integrated with MySQL database and is optimized for underlying infrastructure. You can run analytics on your MySQL data without requiring ETL and without any change to Your applications. Your applications connect to the HeatWave cluster through standard MySQL protocols. The MySQL Database is built on the MySQL Enterprise Edition Server, which allows developers to quickly create and deploy secure cloud native applications using the world's most popular open source database.

This repository contains instructions to run the TPC-C Benchmark™ (TPC-C) with the MySQL HeatWave on OCI or MySQL HeatWave on AWS, using the sysbench framework.

Software prerequisites:

1. Sysbench tpcc benchmark to generate and run TPC-C dataset and queries for workload sizes of your choice
2. MySQL Performance Benchmark kit to get the LUA scripts that enable running sysbench with TPC-C

Required services:

1. Oracle Cloud Infrastructure
2. MySQL HeatWave on OCI or [MySQL HeatWave on AWS]

Getting started

To run TPC-C queries in MySQL HeatWave

1. Install sysbench from the github repository of Akopytov
2. Download TPC-C specific LUA scripts from MySQL Performance Benchmark kit untar the above bundle under directory /BMK create a directory to store the output files

    mkdir -p <yourhomedir>/BMK/BMKouts

3. In TPCC-dim.lua, update the BMK_HOME to point to /BMK

     BMK_HOME = os.getenc("BMK_HOME")
     if (BMK_HOME == nil) then BMK_HOME = "<yourhomedir>/BMK" end

4. Provision a MySQL Database System
5. Assuming you have a running MySQL instance now, create the database. Below sample is for TPCC 100W.

     create database tpcc_100W

6. Create schema under tpcc_100W. Replace the MySQL host IP, username, and password accordingly

   sysbench <yourhomedir>/BMK/sb_exec/lua/TPCC-dim.lua --db-driver=mysql --mysql-storage-engine=InnoDB \  
      --scale=100 --tables=1 --mysql-user=<username> --mysql-password=<passwd> \  
      --mysql-host=<mysql host IP> --mysql-db=tpcc_100W --use_fk=2  --force_null=1 \  
      --events=0 --threads=32 --mysql-ssl=REQUIRED create 2>&1 > <yourhomedir>/BMK/<outfile.out>  

   sysbench <yourhomedir>/BMK/sb_exec/lua/TPCC-dim.lua --db-driver=mysql --mysql-storage-engine=InnoDB \  
      --scale=100 --tables=1 --mysql-user=<username> --mysql-password=<passwd> \  
      --mysql-host=<mysql host IP> --mysql-db=tpcc_100W --use_fk=2  --force_null=1 \  
      --events=0 --threads=32 --mysql-ssl=REQUIRED prepare 2>&1 >> <yourhomedir>/BMK/<outfile.out>

7. Run TPCC workload for varying number of concurrent users using the following snippet

     export BMK_HOME=<yourhomedir>/BMK
     export dt=`date +%Y%m%d%H%M%S`
     export logdir="<yourhomedir>/BMK/BMKouts"
     export duration=300
     i=0

     for users in 128 1 4 16 64 128 256 512 1024 2048 4096
     do
        i=$((i+1))
        logfile=$logdir/${dt}_${users}_${duration}_${i}_out
        sysbench <yourhomedir>/BMK/sb_exec/lua/TPCC-dim.lua --db-driver=mysql --threads=${users} --scale=100 --time=${duration} --warmup-time=30 --report-interval=10 --rand-type=uniform --mysql-host=<mysql host IP> --mysql-user=<username> --mysql-password=<password> --mysql-ssl=REQUIRED --mysql-db=tpcc_100W --tables=1 --events=0 --thread-init-timeout=0 --force_null=1 --rate=0 run 2>&1 >> $logfile
        sleep 15
     done

TPC Benchmark™, QppH, QthH, and QphH are trademarks of the Transaction Processing Performance Council.

All parties are granted permission to copy and distribute to any party without fee all or part of this material provided that: 1) copying and distribution is done for the primary purpose of disseminating TPC material; 2) the TPC copyright notice, the title of the publication, and its date appear, and notice is given that copying is by permission of the Transaction Processing Performance Council.

Benchmark queries are derived from the TPC-C benchmark, but results are not comparable to published TPC-C benchmark results since they do not comply with the TPC-C specification.