pyflink task并行度问题

最新推荐文章于 2024-10-02 10:53:34 发布

scan724

最新推荐文章于 2024-10-02 10:53:34 发布

阅读量520

点赞数 5

分类专栏： Flink实时计算文章标签： python

本文链接：https://blog.csdn.net/zhaoyangjian724/article/details/138498760

版权

Flink实时计算专栏收录该内容

102 篇文章 1 订阅

订阅专栏

from pyflink.datastream import StreamExecutionEnvironment
from pyflink.datastream.functions import RuntimeContext, FlatMapFunction, MapFunction
import json
import re
import logging
import sys
from pyflink.datastream.state import ValueStateDescriptor, MapStateDescriptor
from pyflink.datastream.connectors.kafka import FlinkKafkaConsumer, TypeInformation,FlinkKafkaProducer
from pyflink.common.typeinfo import Types
from pyflink.datastream.connectors.elasticsearch import Elasticsearch7SinkBuilder, ElasticsearchEmitter, FlushBackoffType
from  pyflink.datastream.connectors import  DeliveryGuarantee
from pyflink.common.serialization import SimpleStringSchema
from datetime import datetime




logging.basicConfig(stream=sys.stdout, level=logging.INFO, format="%(asctime)s-%(levelname)s-%(message)s")
logger = logging.getLogger(__name__)

# ���� StreamExecutionEnvironment ����
env = StreamExecutionEnvironment.get_execution_environment()
env.set_parallelism(1)
env.add_jars("file:///root/flink-sql-connector-kafka_2.11-1.14.4.jar")
from pyflink.datastream import DataStream, StreamExecutionEnvironment
from pyflink.datastream.functions import RuntimeContext, FlatMapFunction, MapFunction
from pyflink.common.typeinfo import Types

env = StreamExecutionEnvironment.get_execution_environment()
data = DataStream(env._j_stream_execution_environment.socketTextStream('192.168.137.201', 8899))
#调用map算子,封装成一个task,并行度为8，有8个subtask
ds1=data.map(lambda s: s.upper()).set_parallelism(8)
##sink算子,并行度为4
ds1.print().set_parallelism(4)