Spark Structured Streaming框架(3)之数据输出源详解
2017-09-03 19:58
561 查看
Spark Structured streaming API支持的输出源有:Console、Memory、File和Foreach。其中Console在前两篇博文中已有详述,而Memory使用非常简单。本文着重介绍File和Foreach两种方式,并介绍如何在源码基本扩展新的输出方式。
如下所示的测试例子:
注意:
File形式不能设置"compelete"模型,只能设置"Append"模型。由于Append模型不能有聚合操作,所以将数据保存到外部File时,不能有聚合操作。
上述程序是直接继承ForeachWriter类的接口,并实现了open()、process()、close()三个方法。若采用显示定义一个类来实现,需要注意Scala的泛型设计,如下所示:
如下通过实现一种自定义的Console来介绍这种使用方式:
[2]. Kafka Integration Guide.
1. File
Structured Streaming支持将数据以File形式保存起来,其中支持的文件格式有四种:json、text、csv和parquet。其使用方式也非常简单只需设置checkpointLocation和path即可。checkpointLocation是检查点保存的路径,而path是真实数据保存的路径。如下所示的测试例子:
// Create DataFrame representing the stream of input lines from connection to host:port val lines = spark.readStream .format("socket") .option("host", host) .option("port", port) .load() // Split the lines into words val words = lines.as[String].flatMap(_.split(" ")) // Generate running word count val wordCounts = words.groupBy("value").count() // Start running the query that prints the running counts to the console val query = wordCounts.writeStream .format("json") .option("checkpointLocation","root/jar") .option("path","/root/jar") .start() |
File形式不能设置"compelete"模型,只能设置"Append"模型。由于Append模型不能有聚合操作,所以将数据保存到外部File时,不能有聚合操作。
2. Foreach
foreach输出方式只需要实现ForeachWriter抽象类,并实现三个方法,当Structured Streaming接收到数据就会执行其三个方法,如下的测试示例:/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ // scalastyle:off println package org.apache.spark.examples.sql.streaming import org.apache.spark.sql.SparkSession /** * Counts words in UTF8 encoded, '\n' delimited text received from the network. * * Usage: StructuredNetworkWordCount <hostname> <port> * <hostname> and <port> describe the TCP server that Structured Streaming * would connect to receive data. * * To run this on your local machine, you need to first run a Netcat server * `$ nc -lk 9999` * and then run the example * `$ bin/run-example sql.streaming.StructuredNetworkWordCount * localhost 9999` */ object StructuredNetworkWordCount { def main(args: Array[String]) { if (args.length < 2) { System.err.println("Usage: StructuredNetworkWordCount <hostname> <port>") System.exit(1) } val host = args(0) val port = args(1).toInt val spark = SparkSession .builder .appName("StructuredNetworkWordCount") .getOrCreate() import spark.implicits._ // Create DataFrame representing the stream of input lines from connection to host:port val lines = spark.readStream .format("socket") .option("host", host) .option("port", port) .load() // Start running the query that prints the running counts to the console val query = wordCounts.writeStream .outputMode("append") .foreach(new ForearchWriter[Row]{ override def open(partitionId:Long,version:Long):Boolean={ println("open") return true } override def process(value:Row):Unit={ val spark = SparkSession.builder.getOrCreate() val seq = value.mkString.split(" ") val row = Row.fromSeq(seq) val rowRDD:RDD[Row] = sparkContext.getOrCreate().parallelize[Row](Seq(row)) val userSchema = new StructType().add("name","String").add("age","String") val peopleDF = spark.createDataFrame(rowRDD,userSchema) peopleDF.createOrReplaceTempView(myTable) spark.sql("select * from myTable").show() } override def close(errorOrNull:Throwable):Unit={ println("close") } }) .start() query.awaitTermination() } } // scalastyle:on println |
class myForeachWriter[T<:Row](stream:CatalogTable) extends ForearchWriter[T]{ override def open(partionId:Long,version:Long):Boolean={ println("open") true } override def process(value:T):Unit={ println(value) } override def close(errorOrNull:Throwable):Unit={ println("close") } } |
3. 自定义
若上述Spark Structured Streaming API提供的数据输出源仍不能满足要求,那么还有一种方法可以使用:修改源码。如下通过实现一种自定义的Console来介绍这种使用方式:
3.1 ConsoleSink
Spark有一个Sink接口,用户可以实现该接口的addBatch方法,其中的data参数是接收的数据,如下所示直接将其输出到控制台:class ConsoleSink(streamName:String) extends Sink{ override def addBatch(batchId:Long, data;DataFrame):Unit = { data.show() } } |
3.2 DataStreamWriter
在用户自定义的输出形式时,并调用start()方法后,Spark框架会去调用DataStreamWriter类的start()方法。所以用户可以直接在该方法中添加自定义的输出方式,如我们向其传递上述创建的ConsoleSink类示例,如下所示:def start():StreamingQuery={ if(source == "memory"){ ... }else if(source=="foreach"){ ... }else if(source=="consoleSink"){ val streamName:String = extraOption.get("streamName") mathc{ case Some(str):str case None=>throw new AnalysisException("streamName option must be specified for Sink") } val sink = new consoleSink(streamName) df.sparkSession.sessionState.streamingQueryManager.startQuery( extraOption.get("queryName"), extraOption.get("checkpointLocation"), df, sink, outputMode, useTempCheckpointLocaltion = true, recoverFromCheckpointLocation = false, trigger = trigger ) }else{ ... } } |
3.3 Structured Streaming
在前两部修改和实现完成后,用户就可以按正常的Structured Streaming API方式使用了,唯一不同的是在输出形式传递的参数是"consoleSink"字符串,如下所示:def execute(stream:CatalogTable):Unit={ val spark = SparkSession .builder .appName("StructuredNetworkWordCount") .getOrCreate() /**1. 获取数据对象DataFrame*/ val lines = spark.readStream .format("socket") .option("host", "localhost") .option("port", 9999) .load() /**2. 启动Streaming开始接受数据源的信息*/ val query:StreamingQuery = lines.writeStream .outputMode("append") .format("consoleSink") .option("streamName","myStream") .start() query.awaitTermination() } |
4. 参考文献
[1]. Structured Streaming Programming Guide.[2]. Kafka Integration Guide.
相关文章推荐
- Spark Structured Streaming框架(4)之窗口管理详解
- Spark Structured Streaming框架(5)之进程管理
- Spark Structured Streaming:将数据落地按照数据字段进行分区方案
- 大数据开源框架之二:Berkeley Data Analytics Stack(tachyon, spark, shark, spark streaming)
- Spark Streaming源码解读之流数据不断接收详解
- [spark streaming] ReceiverTracker 数据产生与存储
- 大数据分析处理框架——离线分析(hive,pig,spark)、近似实时分析(Impala)和实时分析(storm、spark streaming)
- Spark-Streaming与Spark-Sql整合实现实时股票排行---通过kafka列队数据
- 大数据IMF传奇行动绝密课程第96课:通过SparkStreaming的foreachRDD把处理后的数据写入外部存储系统中
- Spark Streaming使用Kafka保证数据零丢失
- Spark Streaming 数据接收过程
- Spark Structured Streaming入门编程指南
- 第16课:Spark Streaming源码解读之数据清理内幕彻底解密
- Spark定制班第10课:Spark Streaming源码解读之流数据不断接收全生命周期彻底研究和思考
- 使用Spark Streaming处理爬取的数据(问题总结)
- spark streaming的容错:防止数据丢失
- 大数据框架 Hadoop 和 Spark 的异同
- 大数据计算框架Hadoop, Spark和MPI
- Spark Streaming:大规模流式数据处理的新贵
- Spark Streaming的foreachRDD把处理后的数据写入外部存储系统中