flink dataStream多次sink DAG重复驱动执行?

classic Classic list List threaded Threaded
2 messages Options
lp
Reply | Threaded
Open this post in threaded view
|

flink dataStream多次sink DAG重复驱动执行?

lp

有个疑问,
如下程序片段:

------
Properties properties = new Properties();
        properties.setProperty("bootstrap.servers",kafkaAddr);
        properties.setProperty("group.id",kafkaOdsGroup);
        properties.setProperty("auto.offset.reset",kafkaOdsAutoOffsetReset);
       
properties.setProperty(FlinkKafkaConsumerBase.KEY_PARTITION_DISCOVERY_INTERVAL_MILLIS,kafkaOdsPartitionDiscoverInterval);
       
properties.setProperty("transaction.timeout.ms",KafkaOdsTransactionTimeout);//kafka事务超时时间

        FlinkKafkaConsumer<String> flinkKafkaConsumer = new
FlinkKafkaConsumer<>(kafkaOdsTopic,new SimpleStringSchema(),properties);
        DataStreamSource dataStreamSource =
env.addSource(flinkKafkaConsumer);
        dataStreamSource.printToErr("1");
        dataStreamSource.printToErr("2");
        dataStreamSource.printToErr("3");
----------------


我对一个datastream进行多次相同操作的sink,请问是否会导致上游整个DAG重复驱动执行,基于spark的惯性思维,我认为上游DAG是会重复驱动执行的?



--
Sent from: http://apache-flink.147419.n8.nabble.com/
Reply | Threaded
Open this post in threaded view
|

Re: flink dataStream多次sink DAG重复驱动执行?

Evan
个人理解是不会重复驱动执行的,具体你可以测试一下,但是从底层原理上讲,我也讲不了。
 
发件人: lp
发送时间: 2021-03-05 17:31
收件人: user-zh
主题: flink dataStream多次sink DAG重复驱动执行?
 
有个疑问,
如下程序片段:
 
------
Properties properties = new Properties();
        properties.setProperty("bootstrap.servers",kafkaAddr);
        properties.setProperty("group.id",kafkaOdsGroup);
        properties.setProperty("auto.offset.reset",kafkaOdsAutoOffsetReset);
       
properties.setProperty(FlinkKafkaConsumerBase.KEY_PARTITION_DISCOVERY_INTERVAL_MILLIS,kafkaOdsPartitionDiscoverInterval);
       
properties.setProperty("transaction.timeout.ms",KafkaOdsTransactionTimeout);//kafka事务超时时间
 
        FlinkKafkaConsumer<String> flinkKafkaConsumer = new
FlinkKafkaConsumer<>(kafkaOdsTopic,new SimpleStringSchema(),properties);
        DataStreamSource dataStreamSource =
env.addSource(flinkKafkaConsumer);
        dataStreamSource.printToErr("1");
        dataStreamSource.printToErr("2");
        dataStreamSource.printToErr("3");
----------------
 
 
我对一个datastream进行多次相同操作的sink,请问是否会导致上游整个DAG重复驱动执行,基于spark的惯性思维,我认为上游DAG是会重复驱动执行的?
 
 
 
--
Sent from: http://apache-flink.147419.n8.nabble.com/