Flink SQL Client初探

2020-05-26 14:40:28 浏览数 (1)

关于Flink SQL Client

Flink Table & SQL的API实现了通过SQL语言处理实时技术算业务,但还是要编写部分Java代码(或Scala),并且还要编译构建才能提交到Flink运行环境,这对于不熟悉Java或Scala的开发者就略有些不友好了; SQL Client的目标就是解决上述问题(官方原话with a build tool before being submitted to a cluster.)

局限性

遗憾的是,在Flink-1.10.0版本中,SQL Client只是个Beta版本(不适合用于生产环境),并且只能连接到本地Flink,不能像mysql、cassandra等客户端工具那样远程连接server,这些在将来的版本会解决:

环境信息

接下来采用实战的方式对Flink SQL Client做初步尝试,环境信息如下:

  1. 电脑:MacBook Pro2018 13寸,macOS Catalina 10.15.3
  2. Flink:1.10.0
  3. JDK:1.8.0_211

本地启动flink

  1. 下载flink包,地址:http://ftp.kddilabs.jp/infosystems/apache/flink/flink-1.10.0/flink-1.10.0-bin-scala_2.11.tgz
  2. 解压:tar -zxvf flink-1.10.0-bin-scala_2.11.tgz
  3. 进目录flink-1.10.0/bin/,执行命令./start-cluster.sh启动本地flink;
  4. 访问该机器的8081端口,可见本地flink启动成功:

启动SQL Client CLI

  1. 在目录flink-1.10.0/bin/执行./sql-client.sh即可启动SQL Client CLI,如下图所示,红框中的BETA提醒着在生产环境如果要用此工具:
  1. 第一个要掌握的是HELP命令:
  1. 从hello world开始把,执行命令select ‘Hello world!’;,控制台输出如下图所示,输入Q可退出:

两种展示模式

  1. 第一种是table mode,效果像是对普通数据表的查询,设置该模式的命令:
代码语言:javascript复制
SET execution.result-mode=table;
  1. 第二种是changelog mode,效果像是打印每一次数据变更的日志,设置该模式的命令:
代码语言:javascript复制
SET execution.result-mode=changelog;
  1. 设置table mode后,执行以下命令作一次简单的分组查询:
代码语言:javascript复制
SELECT name, 
  COUNT(*) AS cnt 
  FROM (VALUES ('Bob'), ('Alice'), ('Greg'), ('Bob')) 
  AS NameTable(name) 
  GROUP BY name;
  1. 为了便于对比,下图同时贴上两种模式的查询结果,注意绿框中显示了该行记录是增加还是删除:
  1. 不论是哪种模式,查询结构都保存在SQL Client CLI进程的堆内存中;
  2. 在chenglog模式下,为了保证控制台可以正常输入输出,查询结果只展示最近1000条;
  3. table模式下,可以翻页查询更多结果,结果数量受配置项max-table-result-rows以及可用堆内存限制;

进一步体验

前面写了几行SQL,对Flink SQL Client有了最基本的感受,接下来做进一步的体验,内容如下:

  1. 创建CSV文件,这是个最简单的图书信息表,只有三个字段:名字、数量、类目,一共十条记录;
  2. 创建SQL Client用到的环境配置文件,该文件描述了数据源以及对应的表的信息;
  3. 启动SQL Client,执行SQL查询上述CSV文件;
  4. 整个操作步骤如下图所示:

操作

  1. 首先请确保Flink已经启动;
  2. 创建名为book-store.csv的文件,内容如下:
代码语言:javascript复制
name001,1,aaa
name002,2,aaa
name003,3,bbb
name004,4,bbb
name005,5,bbb
name006,6,ccc
name007,7,ccc
name008,8,ccc
name009,9,ccc
name010,10,ccc
  1. 在flink-1.10.0/conf目录下创建名为book-store.yaml的文件,内容如下:
代码语言:javascript复制
tables:
  - name: BookStore
    type: source-table
    update-mode: append
    connector:
      type: filesystem
      path: "/Users/zhaoqin/temp/202004/26/book-store.csv"
    format:
      type: csv
      fields:
        - name: BookName
          type: VARCHAR
        - name: BookAmount
          type: INT
        - name: BookCatalog
          type: VARCHAR
      line-delimiter: "n"
      comment-prefix: ","
    schema:
      - name: BookName
        type: VARCHAR
      - name: BookAmount
        type: INT
      - name: BookCatalog
        type: VARCHAR
  - name: MyBookView
    type: view
    query: "SELECT BookCatalog, SUM(BookAmount) AS Amount FROM BookStore GROUP BY BookCatalog"


execution:
  planner: blink                    # optional: either 'blink' (default) or 'old'
  type: streaming                   # required: execution mode either 'batch' or 'streaming'
  result-mode: table                # required: either 'table' or 'changelog'
  max-table-result-rows: 1000000    # optional: maximum number of maintained rows in
                                    #   'table' mode (1000000 by default, smaller 1 means unlimited)
  time-characteristic: event-time   # optional: 'processing-time' or 'event-time' (default)
  parallelism: 1                    # optional: Flink's parallelism (1 by default)
  periodic-watermarks-interval: 200 # optional: interval for periodic watermarks (200 ms by default)
  max-parallelism: 16               # optional: Flink's maximum parallelism (128 by default)
  min-idle-state-retention: 0       # optional: table program's minimum idle state time
  max-idle-state-retention: 0       # optional: table program's maximum idle state time

                                    #   (default database of the current catalog by default)
  restart-strategy:                 # optional: restart strategy
    type: fallback                  #   "fallback" to global restart strategy by default

# Configuration options for adjusting and tuning table programs.

# A full list of options and their default values can be found
# on the dedicated "Configuration" page.
configuration:
  table.optimizer.join-reorder-enabled: true
  table.exec.spill-compression.enabled: true
  table.exec.spill-compression.block-size: 128kb

# Properties that describe the cluster to which table programs are submitted to.

deployment:
  response-timeout: 5000
  1. 对于book-store.yaml文件,有以下几处需要注意: a. tables.type等于source-table,表明这是数据源的配置信息; b. tables.connector描述了详细的数据源信息,path是book-store.csv文件的完整路径; c. tables.format描述了文件内容; d. tables.schema描述了数据源表的表结构; ed. type为view表示MyBookView是个视图(参考数据库的视图概念);
  2. 在flink-1.10.0目录执行以下命令,即可启动SQL Client,并指定book-store.yaml为环境配置:
代码语言:javascript复制
bin/sql-client.sh embedded -d conf/book-store.yaml
  1. 查全表:
代码语言:javascript复制
SELECT * FROM BookStore;
  1. 按照BookCatalog分组统计记录数:
代码语言:javascript复制
SELECT BookCatalog, COUNT(*) AS BookCount FROM BookStore GROUP BY BookCatalog;
  1. 查询视图:
代码语言:javascript复制
select * from MyBookView;

至此,Flink SQL Client的初次体验就完成了,咱们此工具算是有了基本了解,接下来的文章会进一步使用Flink SQL Client做些复杂的操作

0 人点赞