edu.iu.dsc.tws.rsched.schedulers.k8s.mpi

Class MPIMasterStarter

java.lang.Object

edu.iu.dsc.tws.rsched.schedulers.k8s.mpi.MPIMasterStarter

public final class MPIMasterStarter
extends java.lang.Object

This class is started in the first pod in a StatefulSet This class will get the PodIP addresses from all pods in a job When getting the IP addresses, it also waits for all pods to become running It saves those IP addresses to hostfile It checks whether password free ssh is enabled between this pod and all other pods in the statefulset It then executes mpirun command to start OpenMPI workers

Method Summary

Modifier and Type	Method and Description
static java.lang.String	commandAsAString(java.lang.String[] commandArray)
static boolean	createHostFile(java.util.ArrayList<java.lang.String> ipList, int workersPerPod) create hostfile for mpirun command first line in the file is the ip of this pod other lines are unordered each line has one ip
static void	executeMpirun(java.lang.String[] command) send mpirun command to shell
static java.lang.String[]	generateCheckSshCommand(java.util.ArrayList<java.lang.String> podIPs)
static java.lang.String[]	generateMPIrunCommand(java.lang.String className, int workersPerPod, java.lang.String jobMasterIP, java.lang.String logPropsFile, java.lang.String jvmMemory)
static void	main(java.lang.String[] args)
static boolean	runScript(java.lang.String[] command) send check ssh script run command to shell

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

main

public static void main(java.lang.String[] args)

createHostFile

public static boolean createHostFile(java.util.ArrayList<java.lang.String> ipList,
                                     int workersPerPod)

create hostfile for mpirun command first line in the file is the ip of this pod other lines are unordered each line has one ip

generateMPIrunCommand

public static java.lang.String[] generateMPIrunCommand(java.lang.String className,
                                                       int workersPerPod,
                                                       java.lang.String jobMasterIP,
                                                       java.lang.String logPropsFile,
                                                       java.lang.String jvmMemory)

executeMpirun

public static void executeMpirun(java.lang.String[] command)

send mpirun command to shell

commandAsAString

public static java.lang.String commandAsAString(java.lang.String[] commandArray)

generateCheckSshCommand

public static java.lang.String[] generateCheckSshCommand(java.util.ArrayList<java.lang.String> podIPs)

runScript

public static boolean runScript(java.lang.String[] command)

send check ssh script run command to shell