Chapter 5. Minimum hardware recommendations

Ceph can run on non-proprietary commodity hardware. Small production clusters and development clusters can run without performance optimization with modest hardware.

Note

Disk space requirements are based on the Ceph daemons' default path under /var/lib/ceph/ directory.

ProcessCriteriaMinimum Recommended

ceph-osd

Processor

1x AMD64 or Intel 64

RAM

For BlueStore OSDs, Red Hat typically recommends a baseline of 16 GB of RAM per OSD host, with an additional 5 GB of RAM per daemon.

OS Disk

1x OS disk per host

Volume Storage

1x storage drive per daemon

block.db

Optional, but Red Hat recommended, 1x SSD or NVMe or Optane partition or logical volume per daemon. Sizing is 4% of block.data for BlueStore

block.wal

Optional, 1x SSD or NVMe or Optane partition or logical volume per daemon. Use a small size, for example 10 GB, and only if it’s faster than the block.db device.

Network

2x 1GB Ethernet NICs

ceph-mon

Processor

1x AMD64 or Intel 64

RAM

1 GB per daemon

Disk Space

15 GB per daemon (50 GB recommended)

Monitor Disk

Optionally,1x SSD disk for leveldb monitor data.

Network

2x 1 GB Ethernet NICs

ceph-mgr

Processor

1x AMD64 or Intel 64

RAM

1 GB per daemon

Network

2x 1 GB Ethernet NICs

ceph-radosgw

Processor

1x AMD64 or Intel 64

RAM

1 GB per daemon

Disk Space

5 GB per daemon

Network

1x 1 GB Ethernet NICs

ceph-mds

Processor

1x AMD64 or Intel 64

RAM

2 GB per daemon

This number is highly dependent on the configurable MDS cache size. The RAM requirement is typically twice as much as the amount set in the mds_cache_memory_limit configuration setting. Note also that this is the memory for your daemon, not the overall system memory.

Disk Space

2 MB per daemon, plus any space required for logging, which might vary depending on the configured log levels.

Network

2x 1 GB Ethernet NICs

Note that this is the same network as the OSDs. If you have a 10 GB network on your OSDs you should use the same on your MDS so that the MDS is not disadvantaged when it comes to latency.