org.apache.hive.spark.client

Interface JobContext

@InterfaceAudience.Private
public interface JobContext

Holds runtime information about the job execution context. An instance of this class is kept on the node hosting a remote Spark context and is made available to jobs being executed via RemoteSparkContext#submit().

Method Summary

Methods

Modifier and Type	Method and Description
List<String>	getAddedJars() Return all added jar path which added through AddJarJob.
File	getLocalTmpDir() Returns a local tmp dir specific to the context
Map<String,List<org.apache.spark.api.java.JavaFutureAction<?>>>	getMonitoredJobs() Return a map from client job Id to corresponding JavaFutureActions.
<T> org.apache.spark.api.java.JavaFutureAction<T>	monitor(org.apache.spark.api.java.JavaFutureAction<T> job, SparkCounters sparkCounters, Set<Integer> cachedRDDIds) Monitor a job.
org.apache.spark.api.java.JavaSparkContext	sc() The shared SparkContext instance.

Method Detail

sc

org.apache.spark.api.java.JavaSparkContext sc()

The shared SparkContext instance.

monitor

<T> org.apache.spark.api.java.JavaFutureAction<T> monitor(org.apache.spark.api.java.JavaFutureAction<T> job,
                                                        SparkCounters sparkCounters,
                                                        Set<Integer> cachedRDDIds)

Monitor a job. This allows job-related information (such as metrics) to be communicated back to the client.

Returns:

The job (unmodified).

getMonitoredJobs

Map<String,List<org.apache.spark.api.java.JavaFutureAction<?>>> getMonitoredJobs()

Return a map from client job Id to corresponding JavaFutureActions.

getAddedJars

List<String> getAddedJars()

Return all added jar path which added through AddJarJob.

getLocalTmpDir

File getLocalTmpDir()

Returns a local tmp dir specific to the context