Git Product home page Git Product logo

ysc / data-generator Goto Github PK

View Code? Open in Web Editor NEW
280.0 24.0 105.0 16.08 MB

如果你在从事大数据BI的工作,想对比一下MySQL、GreenPlum、Elasticsearch、Hive、Spark SQL、Presto、Impala、Drill、HAWQ、Druid、Pinot、Kylin、ClickHouse、Kudu等不同实现方案之间的表现,那你就需要一份标准的数据进行测试,这个开源项目就是为了生成这样的标准数据。

License: Apache License 2.0

Java 100.00%

data-generator's Introduction

数据生成器

如果你在从事大数据BI的工作,想对比一下MySQL、GreenPlum、Elasticsearch、Hive、Spark SQL、Presto、Impala、Drill、HAWQ、Druid、Pinot、Kylin、ClickHouse、Kudu等不同实现方案之间的表现,那你就需要一份标准的数据进行测试,这个开源项目就是为了生成这样的标准数据。

数据模型:src/main/resources/数据模型.png

src/main/resources/数据模型.png

一、编译程序:

 mvn assembly:assembly

二、在MySQL中创建一个数据库,然后执行 src/main/resources/model_ddl.sql 创建对应的表。

三、指定ES的经纬度类型:

curl -H "Content-Type: application/json" -XPUT 'http://192.168.252.193:9200/contract/contract/_bulk' -d '
{ "index":{ "_id": 1} }
{"id":1}
'

curl -H "Content-Type: application/json" -XPUT 'http://192.168.252.193:9200/contract/_mapping/contract' -d '
{
  "properties": {
    "geo_location": {
      "type": "geo_point"
    }
  }
}
'

curl -H "Content-Type: application/json" -XPUT 'http://192.168.252.193:9200/detail/detail/_bulk' -d '
{ "index":{ "_id": 1} }
{"id":1}
'

curl -H "Content-Type: application/json" -XPUT 'http://192.168.252.193:9200/detail/_mapping/detail' -d '
{
  "properties": {
    "geo_location": {
      "type": "geo_point"
    }
  }
}
'

curl -H "Content-Type: application/json" -XPUT 'http://192.168.252.193:9200/area/area/_bulk' -d '
{ "index":{ "_id": 1} }
{"id":1}
'

curl -H "Content-Type: application/json" -XPUT 'http://192.168.252.193:9200/area/_mapping/area' -d '
{
  "properties": {
    "geo_location": {
      "type": "geo_point"
    }
  }
}
'

curl -H "Content-Type: application/json" -XPUT 'http://192.168.252.193:9200/customer/customer/_bulk' -d '
{ "index":{ "_id": 1} }
{"id":1}
'

curl -H "Content-Type: application/json" -XPUT 'http://192.168.252.193:9200/customer/_mapping/customer' -d '
{
  "properties": {
    "geo_location": {
      "type": "geo_point"
    }
  }
}
'

curl -H "Content-Type: application/json" -XPUT 'http://192.168.252.193:9200/sales_staff/sales_staff/_bulk' -d '
{ "index":{ "_id": 1} }
{"id":1}
'

curl -H "Content-Type: application/json" -XPUT 'http://192.168.252.193:9200/sales_staff/_mapping/sales_staff' -d '
{
  "properties": {
    "geo_location": {
      "type": "geo_point"
    }
  }
}
'

四、在当前目录下的config.txt文件中指定配置:

#新增数据是MySQL批量提交记录数量
batchSize=1000
#订单时间开始年份
startYear=2000
#订单时间开始月份
startMonth=1
#订单时间开始天数
startDay=1
#客户数
customerCount=5000
#销售数
salesStaffCount=2000
#合同数
contractCount=20000
#商品数
itemCount=10000
#商品价格上限
priceLimit=1000
#合同最大明细数
contractDetailLimit=100
#合同明细商品最大数量
itemQuantityLimit=100
#将生成的数据保存到哪个MySQL
mysql.url=jdbc:mysql://192.168.252.193:3306/demo?useUnicode=true&characterEncoding=utf8
mysql.user=root
mysql.password=root
mysql.pageSize=10000
#将MySQL里面的数据查出来组装成JSON文档后索引到哪个ES
es.host=192.168.252.193
es.port=9200
#ES批量提交数量
es.batchSize=1000
#可选值为file或者es
#如果选择file,则在当前目录想生成相应的脚本文件,等程序执行完毕后再执行脚本文件将数据索引到ES
#如果选择es,则在数据生成完毕后直接在程序中把数据提交给ES进行索引
es.mode=es
#是否异步多线程的方式进行ES索引
output.async=true
#如果是异步多线程的方式进行ES索引,则需要几个线程
output.async.thread.count=10
#如果ES索引中断,再次索引的时候从哪一页开始索引,0代表第一页
output.start.page=0

五、运行程序:

all in one:

nohup java -Xmx2g -Xms2g -cp data-generator-1.0-jar-with-dependencies.jar org.apdplat.data.generator.Start &

或者

step by step:

1. 生成模拟数据并保存到mysql:

    nohup java -Xmx2g -Xms2g -cp data-generator-1.0-jar-with-dependencies.jar org.apdplat.data.generator.generator.Generator &

2. 将mysql中的数据生成合同文档并提交给ES:

    nohup java -Xmx2g -Xms2g -cp data-generator-1.0-jar-with-dependencies.jar org.apdplat.data.generator.mysql2es.Contract &
    如果es.mode=es则不需要执行如下两步, 只有es.mode=file才需要执行
    chmod +x contract.sh
    nohup ./contract.sh &

3. 将mysql中的数据生成合同明细文档并提交给ES:

    nohup java -Xmx2g -Xms2g -cp data-generator-1.0-jar-with-dependencies.jar org.apdplat.data.generator.mysql2es.ContractDetail &
    如果es.mode=es则不需要执行如下两步, 只有es.mode=file才需要执行
    chmod +x detail.sh
    nohup ./detail.sh &

4. 将mysql中的数据生成区域文档并提交给ES:

    nohup java -Xmx1g -Xms1g -cp data-generator-1.0-jar-with-dependencies.jar org.apdplat.data.generator.mysql2es.Area &
    如果es.mode=es则不需要执行如下两步, 只有es.mode=file才需要执行
    chmod +x area.sh
    nohup ./area.sh &

5. 将mysql中的数据生成商品文档并提交给ES:

    nohup java -Xmx2g -Xms2g -cp data-generator-1.0-jar-with-dependencies.jar org.apdplat.data.generator.mysql2es.Item &
    如果es.mode=es则不需要执行如下两步, 只有es.mode=file才需要执行
    chmod +x item.sh
    nohup ./item.sh &

6. 将mysql中的数据生成客户文档并提交给ES:

    nohup java -Xmx2g -Xms2g -cp data-generator-1.0-jar-with-dependencies.jar org.apdplat.data.generator.mysql2es.Customer &
    如果es.mode=es则不需要执行如下两步, 只有es.mode=file才需要执行
    chmod +x customer.sh
    nohup ./customer.sh &

7. 将mysql中的数据生成销售文档并提交给ES:

    nohup java -Xmx2g -Xms2g -cp data-generator-1.0-jar-with-dependencies.jar org.apdplat.data.generator.mysql2es.SalesStaff &
    如果es.mode=es则不需要执行如下两步, 只有es.mode=file才需要执行
    chmod +x sales_staff.sh
    nohup ./sales_staff.sh &

8. 将mysql中的数据生成品牌文档并提交给ES:

    nohup java -Xmx2g -Xms2g -cp data-generator-1.0-jar-with-dependencies.jar org.apdplat.data.generator.mysql2es.Brand &
    如果es.mode=es则不需要执行如下两步, 只有es.mode=file才需要执行
    chmod +x brand.sh
    nohup ./brand.sh &

9. 将mysql中的数据生成分类文档并提交给ES:

    nohup java -Xmx2g -Xms2g -cp data-generator-1.0-jar-with-dependencies.jar org.apdplat.data.generator.mysql2es.Category &
    如果es.mode=es则不需要执行如下两步, 只有es.mode=file才需要执行
    chmod +x category.sh
    nohup ./category.sh &

六、在hive中执行 src/main/resources/hive_ddl.sql 创建表。

七、执行 src/main/resources/sqoop.txt 中的命令将MySQL中的数据导入Hive。

八、在Kylin中导入Hive的表、创建Model和Cube、构建Cube。

九、在Kibana中创建索引模式,创建图表。

十、对MySQL、Kibana+ES、Kylin进行对比如下统计:

SELECT
    item. NAME ,
    sum(contract_detail.price) AS total_price ,
    sum(contract_detail.item_quantity) AS total_quantity
FROM
    contract_detail
LEFT JOIN item ON contract_detail.item_id = item.id
GROUP BY
    item. NAME
ORDER BY
    total_quantity DESC 
    
Kylin耗时0.5秒,MySQL59秒,ES5秒。

data-generator's People

Contributors

ysc avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.