Installation and Operation Guide for BigBench on MaxCompute Benchmarking Suite

Prerequisites
1、Create An Alibaba Cloud account.
2、The MaxCompute user-end software (preferably 0.27.0 or later) has been installed and configured. For details, see installation and configuration.
3、Java 1.7 (Oracle/OpenJDK) or later has been installed (preferably the 64-bit version).

Download the BigBench suite
1、Click this link to download the BigBench on MaxCompute suite.
2、You can also obtain the BigBench on MaxCompute suite from the GitHub address of the following project.

Configure BigBench on MaxCompute
1、Edit engineSettings.conf in the engines/maxcompute/conf folder and configure the MaxCompute client to run the file as follows.

BINARY="$MAXCOMPUTE_CLIENT_DIR/bin/odpscmd" 

2、Edit userSettings.conf in the conf folder and configure the project name of MaxCompute as follows.

export BIG_BENCH_DEFAULT_DATABASE="$MAXCOMPUTE_PROJECT_NAME"

3、Edit bigBench.properties in the conf folder and configure the test phase of BigBench as follows.

workload=BENCHMARK_START,POWER_TEST,THROUGHPUT_TEST_1,BENCHMARK_STOP,VALIDATE_POWER_TEST,VALIDATE_THROUGHPUT_TEST_1

Use the preceding default test phase unless otherwise specified.

Run the BigBench on MaxCompute benchmarking suite.**
Use the BigBench driver script to run BigBench benchmarking.

$INSTALL_DIR/bin/bigBench runBenchmark

BigBench on MaxCompute log directory
The running log and result log are stored in the following directory:

$INSTALL_DIR/logs

FAQ

How to set MaxCompute parameters in the BigBench benchmarking suite?
Set parameters for all MaxCompute queries in the engineSettings.sql file. The file is located at the following path:

$INSTALL_DIR/engines/maxcompute/conf/engineSettings.sql

The file contains a series of MaxCompute parameters. Use the default setting unless otherwise required.

How to set MaxCompute parameters for a specific query?
Set parameters in the engineLocalSettings.sql file in the queries directory. The file is located at the following path:

$INSTALL_DIR/engines/maxcompute/queries/q??/engineLocalSettings.sql

You can set MaxCompute parameters for a specific query in the preceding file. The BigBench suite automatically imports the file and overwrites the global variables of MaxCompute. Use the default setting unless otherwise required.

Optimizing Complex Data Distribution in MaxCompute For a long time, data distribution has been an issue in the field of Big Data processing. Unfortunately, the Big Data processing systems that are popular today do not satisfactorily solve the issue.
MaxCompute（原ODPS）是一项面向分析的大数据计算服务，它以Serverless架构提供快速、全托管的在线数据仓库服务，消除传统数据平台在资源扩展性和弹性方面的限制，最小化用户运维投入，使您经济并高效的分析处理海量数据。