Client做爲Endpoint的具體實例,下面咱們介紹一下Client啓動以及OnStart指令後的額外工做java
/opt/jdk1.7.0_79/bin/java -cp /opt/spark-2.1.0/conf/:/opt/spark-2.1.0/jars/*:/opt/hadoop-2.6.4/etc/hadoop/ -Xmx1g -XX:MaxPermSize=256m org.apache.spark.deploy.SparkSubmit --master spark://zqh:7077 --class org.apache.spark.examples.SparkPi ../examples/jars/spark-examples_2.11-2.1.0.jar 10
--master]
= spark:* & useRest: org.apache.spark.deploy.rest.RestSubmissionClient--master]
= spark:* & !useRest : org.apache.spark.deploy.Client--master]
= yarn: org.apache.spark.deploy.yarn.Client--master]
= mesos:*: org.apache.spark.deploy.rest.RestSubmissionClient--master]
= spark:* & useRest: 包含primaryResource與mainClass--master]
= spark:* & !useRest : 包含--supervise --memory --cores launch 【childArgs】, primaryResource, mainClass--master]
= yarn:--class --arg --jar/--primary-py-file/--primary-r-file--master]
= mesos:*: primaryResource消息實例 | 發起方 | 接收方 | 說明 |
消息實例 | 發起方 | 接收方 | 說明 |
SubmitDriverResponse | Master | Client | |
KillDriverResponse | Client |