import java.util.ArrayList;
import java.util.List;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.api.java.function.Function2;
import org.apache.spark.api.java.function.PairFlatMapFunction;
import scala.Tuple2;
public class SparkMain {
@SuppressWarnings("serial") public static void main(String[] args) {
SparkConf conf = new SparkConf().setAppName("Spark");
/*独立模式
conf.setMaster("spark://master56:7077");
conf.set("spark.cores.max", "48");
*/
/*yarn-client模式*/
conf.setMaster("yarn-client");
//设置程序包
conf.setJars(new String[]{"/home/hadoop/Spark-0.0.1-SNAPSHOT/lib/Spark-0.0.1-SNAPSHOT.jar"});
//设置SparkHOME
conf.setSparkHome("/home/hadoop/spark-1.2.0-cdh5.3.2");
//设置运行资源参数
conf.set("spark.executor.instances", "30");
conf.set("spark.executor.cores", "3");
conf.set("spark.executor.memory", "5G");
conf.set("spark.driver.memory", "3G");
conf.set("spark.driver.maxResultSize", "10G");
JavaSparkContext context = new JavaSparkContext(conf);
//设置运行资源参数
JavaRDD<String> rdd = context.textFile("hdfs://nujhadoop/spark.txt");
List<Tuple2<String, Integer>> result = rdd.flatMapToPair(new PairFlatMapFunction<String, String, Integer>(){
@Override
public Iterable<Tuple2<String, Integer>> call(String arg0)
throws Exception {
ArrayList<Tuple2<String, Integer>> list = new ArrayList<Tuple2<String, Integer>>();
String[] array = arg0.split(" ");
for (String temper : array) {
list.add(new Tuple2<String, Integer>(temper, 1));
}
return list;
}
}).reduceByKey(new Function2<Integer, Integer, Integer>(){
@Override
public Integer call(Integer arg0, Integer arg1)
throws Exception {
// TODO Auto-generated method stub
return arg0 + arg1;
}
}).collect();
//打印结果
for (Tuple2<String, Integer> temper : result) {
System.out.println(temper._1+","+temper._2);
}
context.stop();
}
}
说明:
FeiNiBukeZyh
- 浏览: 12991 次
- 性别:
- 来自: 沈阳
-
最新评论
一:上传输入文件到hadoop,本例上传的文件名为spark.txt
二:打包程序，打包名为:Spark-0.0.1-SNAPSHOT.jar
三:上传文件到Spark集群进行部署，如:
四:启动程序 sh ./run.sh
日志结果:
Using Spark's default log4j profile: org/apache/spark/log4j-defaults.properties 15/07/15 16:45:10 INFO SecurityManager: Changing view acls to: hadoop 15/07/15 16:45:10 INFO SecurityManager: Changing modify acls to: hadoop 15/07/15 16:45:10 INFO SecurityManager: SecurityManager: authentication disabled; ui acls disabled; users with view permissions: Set(hadoop); users with modify permissions: Set(hadoop) 15/07/15 16:45:11 INFO Slf4jLogger: Slf4jLogger started 15/07/15 16:45:11 INFO Remoting: Starting remoting 15/07/15 16:45:11 INFO Remoting: Remoting started; listening on addresses :[akka.tcp://sparkDriver@slave63:22597] 15/07/15 16:45:11 INFO Remoting: Remoting now listens on addresses: [akka.tcp://sparkDriver@slave63:22597] 15/07/15 16:45:11 INFO Utils: Successfully started service 'sparkDriver' on port 22597. 15/07/15 16:45:11 INFO SparkEnv: Registering MapOutputTracker 15/07/15 16:45:11 INFO SparkEnv: Registering BlockManagerMaster 15/07/15 16:45:11 INFO DiskBlockManager: Created local directory at /tmp/spark-local-20150715164511-17b9 15/07/15 16:45:11 INFO MemoryStore: MemoryStore started with capacity 1635.9 MB 15/07/15 16:45:12 WARN NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable 15/07/15 16:45:12 INFO HttpFileServer: HTTP File server directory is /tmp/spark-bd6a9445-0d51-4d1b-9fc5-b4dcbcdd4cd0 15/07/15 16:45:12 INFO HttpServer: Starting HTTP Server 15/07/15 16:45:12 INFO Utils: Successfully started service 'HTTP file server' on port 54673. 15/07/15 16:45:12 INFO Utils: Successfully started service 'SparkUI' on port 4040. 15/07/15 16:45:12 INFO SparkUI: Started SparkUI at http://slave63:4040 15/07/15 16:45:13 INFO SparkContext: Added JAR /home/hadoop/Spark-0.0.1-SNAPSHOT/lib/Spark-0.0.1-SNAPSHOT.jar at http://172.20.10.63:54673/jars/Spark-0.0.1-SNAPSHOT.jar with timestamp 1436949913052 15/07/15 16:45:13 INFO RMProxy: Connecting to ResourceManager at master46/172.20.10.46:8032 15/07/15 16:45:13 INFO Client: Requesting a new application from cluster with 30 NodeManagers 15/07/15 16:45:13 INFO Client: Verifying our application has not requested more than the maximum memory capability of the cluster (8192 MB per container) 15/07/15 16:45:13 INFO Client: Will allocate AM container, with 3456 MB memory including 384 MB overhead 15/07/15 16:45:13 INFO Client: Setting up container launch context for our AM 15/07/15 16:45:13 INFO Client: Preparing resources for our AM container 15/07/15 16:45:14 INFO Client: Uploading resource file:/home/hadoop/Spark-0.0.1-SNAPSHOT/lib/spark-assembly-1.2.0-cdh5.3.2.jar -> hdfs://nujhadoop/user/hadoop/.sparkStaging/application_1434338096593_8055/spark-assembly-1.2.0-cdh5.3.2.jar 15/07/15 16:45:15 INFO Client: Setting up the launch environment for our AM container 15/07/15 16:45:16 INFO SecurityManager: Changing view acls to: hadoop 15/07/15 16:45:16 INFO SecurityManager: Changing modify acls to: hadoop 15/07/15 16:45:16 INFO SecurityManager: SecurityManager: authentication disabled; ui acls disabled; users with view permissions: Set(hadoop); users with modify permissions: Set(hadoop) 15/07/15 16:45:16 INFO Client: Submitting application 8055 to ResourceManager 15/07/15 16:45:16 INFO YarnClientImpl: Submitted application application_1434338096593_8055 15/07/15 16:45:17 INFO Client: Application report for application_1434338096593_8055 (state: ACCEPTED) 15/07/15 16:45:17 INFO Client: client token: N/A diagnostics: N/A ApplicationMaster host: N/A ApplicationMaster RPC port: -1 queue: root.hadoop start time: 1436949916087 final status: UNDEFINED tracking URL: http://master46:8088/proxy/application_1434338096593_8055/ user: hadoop 15/07/15 16:45:18 INFO Client: Application report for application_1434338096593_8055 (state: ACCEPTED) 15/07/15 16:45:19 INFO Client: Application report for application_1434338096593_8055 (state: ACCEPTED) 15/07/15 16:45:20 INFO Client: Application report for application_1434338096593_8055 (state: ACCEPTED) 15/07/15 16:45:21 INFO Client: Application report for application_1434338096593_8055 (state: ACCEPTED) 15/07/15 16:45:22 INFO Client: Application report for application_1434338096593_8055 (state: ACCEPTED) 15/07/15 16:45:22 INFO YarnClientSchedulerBackend: ApplicationMaster registered as Actor[akka.tcp://sparkYarnAM@slave28:55325/user/YarnAM#945036977] 15/07/15 16:45:22 INFO YarnClientSchedulerBackend: Add WebUI Filter. org.apache.hadoop.yarn.server.webproxy.amfilter.AmIpFilter, Map(PROXY_HOSTS -> master46, PROXY_URI_BASES -> http://master46:8088/proxy/application_1434338096593_8055), /proxy/application_1434338096593_8055 15/07/15 16:45:22 INFO JettyUtils: Adding filter: org.apache.hadoop.yarn.server.webproxy.amfilter.AmIpFilter 15/07/15 16:45:23 INFO Client: Application report for application_1434338096593_8055 (state: RUNNING) 15/07/15 16:45:23 INFO Client: client token: N/A diagnostics: N/A ApplicationMaster host: slave28 ApplicationMaster RPC port: 0 queue: root.hadoop start time: 1436949916087 final status: UNDEFINED tracking URL: http://master46:8088/proxy/application_1434338096593_8055/ user: hadoop 15/07/15 16:45:23 INFO YarnClientSchedulerBackend: Application application_1434338096593_8055 has started running. 15/07/15 16:45:23 INFO NettyBlockTransferService: Server created on 50871 15/07/15 16:45:23 INFO BlockManagerMaster: Trying to register BlockManager 15/07/15 16:45:23 INFO BlockManagerMasterActor: Registering block manager slave63:50871 with 1635.9 MB RAM, BlockManagerId(<driver>, slave63, 50871) 15/07/15 16:45:23 INFO BlockManagerMaster: Registered BlockManager 15/07/15 16:45:28 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave52:23892/user/Executor#469935313] with ID 1 15/07/15 16:45:28 INFO RackResolver: Resolved slave52 to /rack2 15/07/15 16:45:29 INFO BlockManagerMasterActor: Registering block manager slave52:36246 with 2.6 GB RAM, BlockManagerId(1, slave52, 36246) 15/07/15 16:45:33 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave63:19749/user/Executor#-1474529488] with ID 4 15/07/15 16:45:33 INFO RackResolver: Resolved slave63 to /rack2 15/07/15 16:45:34 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave28:51624/user/Executor#1260742083] with ID 28 15/07/15 16:45:34 INFO RackResolver: Resolved slave28 to /rack3 15/07/15 16:45:34 INFO BlockManagerMasterActor: Registering block manager slave63:64068 with 2.6 GB RAM, BlockManagerId(4, slave63, 64068) 15/07/15 16:45:35 INFO BlockManagerMasterActor: Registering block manager slave28:17967 with 2.6 GB RAM, BlockManagerId(28, slave28, 17967) 15/07/15 16:45:36 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave23:57756/user/Executor#-1426187042] with ID 16 15/07/15 16:45:36 INFO RackResolver: Resolved slave23 to /rack3 15/07/15 16:45:37 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave36:35348/user/Executor#-1773874771] with ID 3 15/07/15 16:45:37 INFO RackResolver: Resolved slave36 to /rack1 15/07/15 16:45:37 INFO BlockManagerMasterActor: Registering block manager slave23:62605 with 2.6 GB RAM, BlockManagerId(16, slave23, 62605) 15/07/15 16:45:38 INFO BlockManagerMasterActor: Registering block manager slave36:23663 with 2.6 GB RAM, BlockManagerId(3, slave36, 23663) 15/07/15 16:45:39 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave15:43551/user/Executor#-576231312] with ID 11 15/07/15 16:45:39 INFO RackResolver: Resolved slave15 to /rack3 15/07/15 16:45:40 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave37:40681/user/Executor#-1501756719] with ID 29 15/07/15 16:45:40 INFO RackResolver: Resolved slave37 to /rack1 15/07/15 16:45:40 INFO BlockManagerMasterActor: Registering block manager slave15:55745 with 2.6 GB RAM, BlockManagerId(11, slave15, 55745) 15/07/15 16:45:41 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave26:28665/user/Executor#1165917342] with ID 21 15/07/15 16:45:41 INFO RackResolver: Resolved slave26 to /rack3 15/07/15 16:45:41 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave54:37653/user/Executor#587407704] with ID 2 15/07/15 16:45:41 INFO RackResolver: Resolved slave54 to /rack2 15/07/15 16:45:41 INFO BlockManagerMasterActor: Registering block manager slave37:38747 with 2.6 GB RAM, BlockManagerId(29, slave37, 38747) 15/07/15 16:45:42 INFO BlockManagerMasterActor: Registering block manager slave26:46197 with 2.6 GB RAM, BlockManagerId(21, slave26, 46197) 15/07/15 16:45:42 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave17:64410/user/Executor#-1365579611] with ID 19 15/07/15 16:45:42 INFO RackResolver: Resolved slave17 to /rack3 15/07/15 16:45:42 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave35:15510/user/Executor#972094812] with ID 15 15/07/15 16:45:42 INFO RackResolver: Resolved slave35 to /rack1 15/07/15 16:45:42 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave55:36974/user/Executor#-597250789] with ID 26 15/07/15 16:45:42 INFO RackResolver: Resolved slave55 to /rack2 15/07/15 16:45:42 INFO BlockManagerMasterActor: Registering block manager slave54:18807 with 2.6 GB RAM, BlockManagerId(2, slave54, 18807) 15/07/15 16:45:43 INFO YarnClientSchedulerBackend: SchedulerBackend is ready for scheduling beginning after waiting maxRegisteredResourcesWaitingTime: 30000(ms) 15/07/15 16:45:43 INFO BlockManagerMasterActor: Registering block manager slave17:58808 with 2.6 GB RAM, BlockManagerId(19, slave17, 58808) 15/07/15 16:45:43 INFO BlockManagerMasterActor: Registering block manager slave35:29737 with 2.6 GB RAM, BlockManagerId(15, slave35, 29737) 15/07/15 16:45:43 INFO MemoryStore: ensureFreeSpace(261904) called with curMem=0, maxMem=1715396935 15/07/15 16:45:43 INFO MemoryStore: Block broadcast_0 stored as values in memory (estimated size 255.8 KB, free 1635.7 MB) 15/07/15 16:45:43 INFO BlockManagerMasterActor: Registering block manager slave55:29257 with 2.6 GB RAM, BlockManagerId(26, slave55, 29257) 15/07/15 16:45:43 INFO MemoryStore: ensureFreeSpace(21065) called with curMem=261904, maxMem=1715396935 15/07/15 16:45:43 INFO MemoryStore: Block broadcast_0_piece0 stored as bytes in memory (estimated size 20.6 KB, free 1635.7 MB) 15/07/15 16:45:43 INFO BlockManagerInfo: Added broadcast_0_piece0 in memory on slave63:50871 (size: 20.6 KB, free: 1635.9 MB) 15/07/15 16:45:43 INFO BlockManagerMaster: Updated info of block broadcast_0_piece0 15/07/15 16:45:43 INFO SparkContext: Created broadcast 0 from textFile at SparkMain.java:31 15/07/15 16:45:44 INFO FileInputFormat: Total input paths to process : 1 15/07/15 16:45:44 INFO SparkContext: Starting job: collect at SparkMain.java:53 15/07/15 16:45:44 INFO DAGScheduler: Registering RDD 2 (flatMapToPair at SparkMain.java:32) 15/07/15 16:45:44 INFO DAGScheduler: Got job 0 (collect at SparkMain.java:53) with 2 output partitions (allowLocal=false) 15/07/15 16:45:44 INFO DAGScheduler: Final stage: Stage 1(collect at SparkMain.java:53) 15/07/15 16:45:44 INFO DAGScheduler: Parents of final stage: List(Stage 0) 15/07/15 16:45:44 INFO DAGScheduler: Missing parents: List(Stage 0) 15/07/15 16:45:44 INFO DAGScheduler: Submitting Stage 0 (FlatMappedRDD[2] at flatMapToPair at SparkMain.java:32), which has no missing parents 15/07/15 16:45:44 INFO MemoryStore: ensureFreeSpace(3672) called with curMem=282969, maxMem=1715396935 15/07/15 16:45:44 INFO MemoryStore: Block broadcast_1 stored as values in memory (estimated size 3.6 KB, free 1635.7 MB) 15/07/15 16:45:44 INFO MemoryStore: ensureFreeSpace(2190) called with curMem=286641, maxMem=1715396935 15/07/15 16:45:44 INFO MemoryStore: Block broadcast_1_piece0 stored as bytes in memory (estimated size 2.1 KB, free 1635.7 MB) 15/07/15 16:45:44 INFO BlockManagerInfo: Added broadcast_1_piece0 in memory on slave63:50871 (size: 2.1 KB, free: 1635.9 MB) 15/07/15 16:45:44 INFO BlockManagerMaster: Updated info of block broadcast_1_piece0 15/07/15 16:45:44 INFO SparkContext: Created broadcast 1 from broadcast at DAGScheduler.scala:838 15/07/15 16:45:44 INFO DAGScheduler: Submitting 2 missing tasks from Stage 0 (FlatMappedRDD[2] at flatMapToPair at SparkMain.java:32) 15/07/15 16:45:44 INFO YarnClientClusterScheduler: Adding task set 0.0 with 2 tasks 15/07/15 16:45:44 INFO RackResolver: Resolved slave38 to /rack1 15/07/15 16:45:44 INFO TaskSetManager: Starting task 0.0 in stage 0.0 (TID 0, slave63, NODE_LOCAL, 1340 bytes) 15/07/15 16:45:44 INFO TaskSetManager: Starting task 1.0 in stage 0.0 (TID 1, slave63, NODE_LOCAL, 1340 bytes) 15/07/15 16:45:45 INFO BlockManagerInfo: Added broadcast_1_piece0 in memory on slave63:64068 (size: 2.1 KB, free: 2.6 GB) 15/07/15 16:45:45 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave65:57998/user/Executor#-1382810865] with ID 12 15/07/15 16:45:45 INFO RackResolver: Resolved slave65 to /rack2 15/07/15 16:45:45 INFO BlockManagerInfo: Added broadcast_0_piece0 in memory on slave63:64068 (size: 20.6 KB, free: 2.6 GB) 15/07/15 16:45:46 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave53:59085/user/Executor#-1064055348] with ID 13 15/07/15 16:45:46 INFO RackResolver: Resolved slave53 to /rack2 15/07/15 16:45:46 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave62:24319/user/Executor#-139207262] with ID 14 15/07/15 16:45:46 INFO RackResolver: Resolved slave62 to /rack2 15/07/15 16:45:46 INFO BlockManagerMasterActor: Registering block manager slave65:64372 with 2.6 GB RAM, BlockManagerId(12, slave65, 64372) 15/07/15 16:45:47 INFO BlockManagerMasterActor: Registering block manager slave62:53823 with 2.6 GB RAM, BlockManagerId(14, slave62, 53823) 15/07/15 16:45:47 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave32:28461/user/Executor#-2071109973] with ID 20 15/07/15 16:45:47 INFO RackResolver: Resolved slave32 to /rack1 15/07/15 16:45:47 INFO BlockManagerMasterActor: Registering block manager slave53:60055 with 2.6 GB RAM, BlockManagerId(13, slave53, 60055) 15/07/15 16:45:47 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave14:35963/user/Executor#148583350] with ID 22 15/07/15 16:45:47 INFO RackResolver: Resolved slave14 to /rack3 15/07/15 16:45:48 INFO BlockManagerMasterActor: Registering block manager slave32:35445 with 2.6 GB RAM, BlockManagerId(20, slave32, 35445) 15/07/15 16:45:48 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave43:63661/user/Executor#1541284948] with ID 24 15/07/15 16:45:48 INFO RackResolver: Resolved slave43 to /rack1 15/07/15 16:45:48 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave33:46267/user/Executor#-1437439698] with ID 10 15/07/15 16:45:48 INFO RackResolver: Resolved slave33 to /rack1 15/07/15 16:45:48 INFO BlockManagerMasterActor: Registering block manager slave43:34953 with 2.6 GB RAM, BlockManagerId(24, slave43, 34953) 15/07/15 16:45:49 INFO BlockManagerMasterActor: Registering block manager slave14:53473 with 2.6 GB RAM, BlockManagerId(22, slave14, 53473) 15/07/15 16:45:49 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave42:26170/user/Executor#794862330] with ID 5 15/07/15 16:45:49 INFO RackResolver: Resolved slave42 to /rack1 15/07/15 16:45:49 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave44:35394/user/Executor#1035079905] with ID 18 15/07/15 16:45:49 INFO RackResolver: Resolved slave44 to /rack1 15/07/15 16:45:49 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave16:52328/user/Executor#1181615525] with ID 30 15/07/15 16:45:49 INFO RackResolver: Resolved slave16 to /rack3 15/07/15 16:45:49 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave13:48403/user/Executor#-1103053012] with ID 27 15/07/15 16:45:49 INFO RackResolver: Resolved slave13 to /rack3 15/07/15 16:45:49 INFO BlockManagerMasterActor: Registering block manager slave42:60923 with 2.6 GB RAM, BlockManagerId(5, slave42, 60923) 15/07/15 16:45:50 INFO BlockManagerMasterActor: Registering block manager slave44:30133 with 2.6 GB RAM, BlockManagerId(18, slave44, 30133) 15/07/15 16:45:50 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave45:63922/user/Executor#-917535710] with ID 6 15/07/15 16:45:50 INFO RackResolver: Resolved slave45 to /rack1 15/07/15 16:45:50 INFO BlockManagerMasterActor: Registering block manager slave16:21970 with 2.6 GB RAM, BlockManagerId(30, slave16, 21970) 15/07/15 16:45:50 INFO BlockManagerMasterActor: Registering block manager slave13:57504 with 2.6 GB RAM, BlockManagerId(27, slave13, 57504) 15/07/15 16:45:50 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave25:18514/user/Executor#-799832935] with ID 25 15/07/15 16:45:50 INFO RackResolver: Resolved slave25 to /rack3 15/07/15 16:45:51 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave27:64380/user/Executor#-520443684] with ID 9 15/07/15 16:45:51 INFO RackResolver: Resolved slave27 to /rack3 15/07/15 16:45:51 INFO BlockManagerMasterActor: Registering block manager slave25:16330 with 2.6 GB RAM, BlockManagerId(25, slave25, 16330) 15/07/15 16:45:51 INFO BlockManagerMasterActor: Registering block manager slave45:63841 with 2.6 GB RAM, BlockManagerId(6, slave45, 63841) 15/07/15 16:45:51 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave24:46357/user/Executor#1463308812] with ID 8 15/07/15 16:45:51 INFO RackResolver: Resolved slave24 to /rack3 15/07/15 16:45:51 INFO TaskSetManager: Finished task 1.0 in stage 0.0 (TID 1) in 7633 ms on slave63 (1/2) 15/07/15 16:45:51 INFO BlockManagerMasterActor: Registering block manager slave33:50916 with 2.6 GB RAM, BlockManagerId(10, slave33, 50916) 15/07/15 16:45:52 INFO TaskSetManager: Finished task 0.0 in stage 0.0 (TID 0) in 7804 ms on slave63 (2/2) 15/07/15 16:45:52 INFO DAGScheduler: Stage 0 (flatMapToPair at SparkMain.java:32) finished in 7.810 s 15/07/15 16:45:52 INFO YarnClientClusterScheduler: Removed TaskSet 0.0, whose tasks have all completed, from pool 15/07/15 16:45:52 INFO DAGScheduler: looking for newly runnable stages 15/07/15 16:45:52 INFO DAGScheduler: running: Set() 15/07/15 16:45:52 INFO DAGScheduler: waiting: Set(Stage 1) 15/07/15 16:45:52 INFO DAGScheduler: failed: Set() 15/07/15 16:45:52 INFO DAGScheduler: Missing parents for Stage 1: List() 15/07/15 16:45:52 INFO DAGScheduler: Submitting Stage 1 (ShuffledRDD[3] at reduceByKey at SparkMain.java:44), which is now runnable 15/07/15 16:45:52 INFO MemoryStore: ensureFreeSpace(2232) called with curMem=288831, maxMem=1715396935 15/07/15 16:45:52 INFO MemoryStore: Block broadcast_2 stored as values in memory (estimated size 2.2 KB, free 1635.7 MB) 15/07/15 16:45:52 INFO MemoryStore: ensureFreeSpace(1403) called with curMem=291063, maxMem=1715396935 15/07/15 16:45:52 INFO MemoryStore: Block broadcast_2_piece0 stored as bytes in memory (estimated size 1403.0 B, free 1635.7 MB) 15/07/15 16:45:52 INFO BlockManagerInfo: Added broadcast_2_piece0 in memory on slave63:50871 (size: 1403.0 B, free: 1635.9 MB) 15/07/15 16:45:52 INFO BlockManagerMaster: Updated info of block broadcast_2_piece0 15/07/15 16:45:52 INFO SparkContext: Created broadcast 2 from broadcast at DAGScheduler.scala:838 15/07/15 16:45:52 INFO DAGScheduler: Submitting 2 missing tasks from Stage 1 (ShuffledRDD[3] at reduceByKey at SparkMain.java:44) 15/07/15 16:45:52 INFO YarnClientClusterScheduler: Adding task set 1.0 with 2 tasks 15/07/15 16:45:52 INFO TaskSetManager: Starting task 0.0 in stage 1.0 (TID 2, slave63, PROCESS_LOCAL, 1121 bytes) 15/07/15 16:45:52 INFO TaskSetManager: Starting task 1.0 in stage 1.0 (TID 3, slave26, PROCESS_LOCAL, 1121 bytes) 15/07/15 16:45:52 INFO BlockManagerInfo: Added broadcast_2_piece0 in memory on slave63:64068 (size: 1403.0 B, free: 2.6 GB) 15/07/15 16:45:52 INFO MapOutputTrackerMasterActor: Asked to send map output locations for shuffle 0 to sparkExecutor@slave63:19749 15/07/15 16:45:52 INFO MapOutputTrackerMaster: Size of output statuses for shuffle 0 is 147 bytes 15/07/15 16:45:52 INFO BlockManagerMasterActor: Registering block manager slave27:35965 with 2.6 GB RAM, BlockManagerId(9, slave27, 35965) 15/07/15 16:45:52 INFO TaskSetManager: Finished task 0.0 in stage 1.0 (TID 2) in 159 ms on slave63 (1/2) 15/07/15 16:45:52 INFO YarnClientSchedulerBackend: Registered executor: Actor[akka.tcp://sparkExecutor@slave18:54423/user/Executor#495118309] with ID 7 15/07/15 16:45:52 INFO RackResolver: Resolved slave18 to /rack3 15/07/15 16:45:52 INFO BlockManagerMasterActor: Registering block manager slave24:57590 with 2.6 GB RAM, BlockManagerId(8, slave24, 57590) 15/07/15 16:45:53 INFO BlockManagerMasterActor: Registering block manager slave18:51244 with 2.6 GB RAM, BlockManagerId(7, slave18, 51244) 15/07/15 16:45:53 INFO BlockManagerInfo: Added broadcast_2_piece0 in memory on slave26:46197 (size: 1403.0 B, free: 2.6 GB) 15/07/15 16:45:53 INFO MapOutputTrackerMasterActor: Asked to send map output locations for shuffle 0 to sparkExecutor@slave26:28665 15/07/15 16:45:53 INFO TaskSetManager: Finished task 1.0 in stage 1.0 (TID 3) in 1605 ms on slave26 (2/2) 15/07/15 16:45:53 INFO DAGScheduler: Stage 1 (collect at SparkMain.java:53) finished in 1.612 s 15/07/15 16:45:53 INFO YarnClientClusterScheduler: Removed TaskSet 1.0, whose tasks have all completed, from pool 15/07/15 16:45:53 INFO DAGScheduler: Job 0 finished: collect at SparkMain.java:53, took 9.550722 s So,1 up.He,1 are,1 got,1 decided,1 bunch,1 his,1 few,1 away,1 backed,1 said��I,1 They,1 air,,1 ripe,1 am,1 never,1 One,1 tried,1 last,1 feeling,1 with,1 day,1 start,,1 One,,1 again,,2 paces,,1 three,,1 they,1 just,1 again,1 still,,1 two,,1 grapes.,1 walked,2 summer,1 walking,1 running,1 up,2 not,1 it,1 He,1 fox,2 orchard.,1 succeeded.,1 was,1 sour.��,1 grapes.The,1 a,4 stopped,1 nose,1 At,1 missed,1 before,1 to,1 back.,1 sure,1 he,5 through,1 thirsty,",1 in,1 could,1 grapes.He,1 of,1 hot,1 juicy."I'm,1 were,1 reach,1 an,1 but,3 jumped,2 and,3 up,,1 give,1 thought.,1 the,3 15/07/15 16:45:53 INFO SparkUI: Stopped Spark web UI at http://slave63:4040 15/07/15 16:45:53 INFO DAGScheduler: Stopping DAGScheduler 15/07/15 16:45:53 INFO YarnClientSchedulerBackend: Shutting down all executors 15/07/15 16:45:53 INFO YarnClientSchedulerBackend: Asking each executor to shut down 15/07/15 16:45:53 INFO YarnClientSchedulerBackend: Stopped 15/07/15 16:45:54 INFO MapOutputTrackerMasterActor: MapOutputTrackerActor stopped! 15/07/15 16:45:54 INFO MemoryStore: MemoryStore cleared 15/07/15 16:45:54 INFO BlockManager: BlockManager stopped 15/07/15 16:45:54 INFO BlockManagerMaster: BlockManagerMaster stopped 15/07/15 16:45:54 INFO SparkContext: Successfully stopped SparkContext 15/07/15 16:45:54 INFO RemoteActorRefProvider$RemotingTerminator: Shutting down remote daemon.
分享到：
发表评论
相关推荐
Spark on Yarn 模式部署是一种常见的 Spark 应用场景，本文将详细介绍 Spark on Yarn 模式部署的步骤和配置过程。 标题解释 Spark on Yarn 模式部署是指将 Spark 应用程序部署在 Yarn 集群上，使得 Spark 能够使用 ...
Spark on Yarn是一种将Spark应用程序部署在Hadoop YARN资源管理器上的方法，它允许Spark充分利用YARN的资源管理和调度功能。在这个实验中，我们将详细探讨如何在Yarn模式下安装和部署Spark集群。 首先，我们需要...
此时，Yarn（Yet Another Resource Negotiator）出现了，它是Hadoop 2.0中引入的新资源管理框架，进一步优化了资源管理和任务调度，而Spark on Yarn模式则充分利用了Yarn的资源调度能力和Spark内存计算的优势，显著...
**Spark on Yarn** 是指Apache Spark集群管理器与Apache Hadoop YARN资源管理器之间的集成模式。这种模式下，YARN作为资源调度器负责分配资源，而Spark则负责任务的具体执行。这种方式使得Spark能够更好地利用Hadoop...
在大数据处理领域，Hadoop MapReduce 和 Apache Spark 是两种重要的计算框架，它们都在YARN（Yet Another Resource Negotiator）上运行以实现资源管理和任务调度。本文将深入探讨这两个框架以及YARN的相关概念。 ...
Spark on YARN 集群搭建详细过程 _title_：Spark on YARN 集群搭建详细过程 _description_：本文详细介绍了 Spark on YARN 集群搭建的过程，包括配置主机 hosts 文件、免密码登录、安装 Java、安装 Scala 等步骤。...
Spark 初始化源码阅读 Spark on YARN 的 Client 和 Cluster 区别 Spark 是一个大数据处理的开源框架，它可以在多种集群管理器上运行，如 YARN、Mesos 和 Standalone。Spark on YARN 是 Spark 在 YARN 集群管理器上...
2. 修改 `conf/spark-defaults.conf`，设置 Spark 在 YARN 上运行的相关参数，如 `spark.master` 设为 `yarn`，`spark.executor.instances` 表示执行器实例数量等。 3. 修改 `conf/spark-env.sh`，设置环境变量，如 ...
Spark on Yan集群搭建的详细过程，减少集群搭建的时间
基于docker搭建spark on yarn及可视化桌面.doc
【Spark on Yarn之Executor内存管理】 Spark是一个分布式计算框架，它可以在多个集群管理系统上运行，包括YARN（Hadoop的资源调度器）。Executor是Spark执行任务的基本单元，每个Executor在一个工作节点上运行，...
Spark on YARN 集群搭建是一个复杂的过程，涉及到多台服务器的配置和软件的安装。以下是详细步骤和相关知识点： 1. **主机配置与网络通信** - `/etc/hosts` 文件配置至关重要，它用于解析主机名到IP地址的映射。...
Spark的Yarn模式是将Spark应用部署在Hadoop的YARN资源管理系统上的方式，这种方式无需单独搭建Spark集群，而是利用YARN的资源管理和调度能力。YARN（Yet Another Resource Negotiator）是Hadoop 2.x版本引入的一个...
基于SparkonYarn的淘宝数据挖掘平台
SPARK2_ON_YARN-2.4.0 jar包下载
本来不打算写的了，但是真的是闲来无事，整天看美剧也没啥意思。这一章打算讲一下Spark onyarn的实现，1.0.0里面...在第一章《spark-submit提交作业过程》的时候，我们讲过Sparkonyarn的在cluster模式下它的main clas
在提交Spark任务前，需要配置Spark的相关属性，如`spark.master`设置为`yarn-client`或`yarn-cluster`，前者用于客户端模式，后者用于集群模式。此外，还需指定Hadoop的配置目录，例如`spark.yarn.conf.archive`。 ...
### 基于Spark_on_Yarn的淘宝数据挖掘平台 #### 一、为什么选择Spark_on_Yarn 在大数据处理领域，随着数据量的急剧增长和技术的发展，传统的数据处理框架如Hadoop MapReduce面临着一系列挑战。淘宝作为中国最大的...
三种方式的spark on kubernetes对比，第一种：spark原生支持Kubernetes资源调度；第二种：google集成的Kubernetes的spark插件sparkoperator；第三种：standalone方式运行spark集群
Spark on Yarn是指在Yarn集群中运行Spark应用程序。下面是Spark on Yarn的配置步骤： 1. 修改配置文件：修改spark-defaults.conf和spark-env.sh文件，配置Spark的参数。 2. 提交Spark任务：使用spark-submit命令...