Module datarush.hadoop.core
Class YarnClusterJobExecutor
- java.lang.Object
-
- com.pervasive.datarush.cal.dr.AbstractDRJobExecutor
-
- com.actian.dataflow.cal.yarn.client.YarnClusterJobExecutor
-
- All Implemented Interfaces:
JobExecutorService
public class YarnClusterJobExecutor extends com.pervasive.datarush.cal.dr.AbstractDRJobExecutor
A job executor is created for each new job. The executor wraps several pieces of needed information including the file client, job master (started with thestart(FileClient, JobSpecifier)
method), and the job specifier (which itself contains the classpath specifier, job info, requested resources, and storage management path for the job).
-
-
Constructor Summary
Constructors Constructor Description YarnClusterJobExecutor(FileClient fileClient, JobSpecifier jobSpec)
-
Method Summary
All Methods Instance Methods Concrete Methods Modifier and Type Method Description StorageManager
getExternalStorageManager()
Has the side effect of starting the application master.protected com.pervasive.datarush.cal.dr.AbstractDRJobExecutor.JobMasterHelper
start(FileClient fileClient, JobSpecifier jobSpec)
-
-
-
Constructor Detail
-
YarnClusterJobExecutor
public YarnClusterJobExecutor(FileClient fileClient, JobSpecifier jobSpec)
-
-
Method Detail
-
getExternalStorageManager
public StorageManager getExternalStorageManager()
Has the side effect of starting the application master.- Returns:
- the storage manager
-
start
protected com.pervasive.datarush.cal.dr.AbstractDRJobExecutor.JobMasterHelper start(FileClient fileClient, JobSpecifier jobSpec)
- Specified by:
start
in classcom.pervasive.datarush.cal.dr.AbstractDRJobExecutor
-
-