You should go through some more articles regarding cluster to make better understanding.
I have put here the steps of taking cluster backup and restore with 4 data node scenario.
1.MySQL Cluster Native Backup Tool (ONLINE Backup) and Restore
ndb_mgm> START BACKUP <Option>
<Option>:NOWAIT, WAIT STARTED, WAIT COMPLETED
NOWAIT - This returns a response to the display immediately and begins the backup without waiting for responses from the data nodes.
WAIT STARTED - This returns a response to the display only when a response is returned from all the data nodes saying the backup has begun.
WAIT COMPLETED - This returns a response to the display only when a response is returned from all the data nodes saying that the backups have been completed successfully.
In most cases, using the START BACKUP WAIT COMPLETED is the best option:
ndb_mgm> START BACKUP WAIT COMPLETED
Files that are created on Disk
1.1.1 .ctl - This group of files (labeled backup-..ctl) is the metadata from the cluster for each data node.
1.1.2 .log - This group of files (labeled backup-..log) contain the UNDO log files to keep the backup consistent and to ensure that no locks are needed during the backup process.
1.1.3 .Data - This group of files (labeled backup-- ..Data) contains the data for the entire system.
1.2 To interrupt the backup in progress.
If the abort was successful, a message will be displayed stating this and issuing an error (which is to be expected).If a backup id is entered, but there is no backup occurring that is associated with that id, there is a response but without an error being issued:
ndb_mgm> ABORT BACKUP <backup id>
ndb_mgm> ABORT BACKUP 6
ndb_mgm> <node_id> REPORT BackupStatus
This command provides a status report on a cluster backup at anytime during the backup for node_id, or for all data nodes using ALL.
1.4 Restoring cluster Backup
1.4.1 Shutdown the cluster and then Restart Management Node
1.4.2 Start and empty the data nodes:-
Prior to restoring a backup, it is important to start and empty the data nodes.This is accomplished by starting each data node with the --initial option.
1.4.3 Restore the Cluster Backup
ndb_mgm> ENTER SINGLE USER MODE [Node-Id of First Free Slot]
18.104.22.168 In an operating system console, enter the following command from the /var/lib/mysqlcluster/BACKUP/BACKUP-1/ directory:
ndb_restore -m -r -b 1 -n 2 /* -- For First DataNode*/
ndb_restore -r -b 1 -n 3 /*-- For Second DataNode */
ndb_mgm> EXIT SINGLE USER MODE
2. MySQL Cluster Backup through MySQLDump Utility
shell> mysqldump -u root -p --lock-tables Dd_Name > Db_Name-backup.sql
2.2 Restoring mysqldump Files
shell> mysql -u root -p < Db_Name-backup.sql