Review the recommended layout for master and worker nodes.
The following layout is recommended for master nodes:
- 2 x Disks (capacity at least 500 GB) in RAID 1 (software or hardware) for operating system (OS) and logs.
- 4 x Disks (>= 1 TB each) in RAID 10 for Database data (see Note).
- 2 x Disks (capacity at least 1 TB) in RAID 1 (software or hardware) for NameNode metadata.
- 1 x Disk JBOD/RAID 0 for ZooKeeper (>= 1TB) (see Note).
- 1 x Disk JBOD/RAID 0 for Quorum JournalNode (>= 1 TB).
The following layout is recommended for worker nodes:
- 2x Disks (capacity at least 500 GB) in RAID 1 (software or hardware) for OS and logs.
- 12-24 SATA Disks JBOD mode (or as multiple single-drive RAID 0 volumes if using a RAID controller incapable of doing JBOD passthrough) no larger than 4 TB in capacity. If the RAID controller has cache, use it for write caching (preferably with battery backup) and disable read caching. Follow your hardware vendor’s best practices where available.
- For a higher performance profile, use 10K RPM SATA or faster SAS drives. These often have lower capacity, but capacity considerations can be offset by adding more data nodes.
RAID controllers should be configured to disable any optimization settings for the RAID 0 volumes.