We decided to create a cluster with the five VMs we had at disposition. They are the following:
-
MAPD-B_Gr09-5 10.67.22.226
-
MAPD-B_Gr09-4 10.67.22.68
-
MAPD-B_Gr09-3 10.67.22.63
-
MAPD-B_Gr09-2 10.67.22.88
-
MAPD-B_Gr09-1 10.67.22.100
In particular, we set the last one as the master node, and the others as pure workers.
Once you arrive on MAPD-B_Gr09-1 (master) type cd ../home/MAPD_project and ./start-all.sh to start either the Kafka Zookeper, Server and all Spark cluster infrastructure.