Unlike in filestore, where every write is written in its entirety to both the journal and then finally to disk, in BlueStore, the data part of the write in most cases is written directly to the block device. This removes the double-write penalty and, on pure spinning-disk OSDs, dramatically improves performance and lowers SSD wear. However, as mentioned previously, this double write has a positive side effect of decreasing write latency when the spinning disks are combined with SSD journals. BlueStore can also use flash-based storage devices to lower write latency by deferring writes, first writing data into the RocksDB WAL and then later flushing these entries to disk. Unlike filestore, not every write is written into the WAL; configuration parameters determine the I/O size cut-off as to what writes are deferred. The configuration parameter is shown in the following code:

bluestore_prefer_deferred_size

This controls the size of I/Os that will be written to the WAL first. For spinning disks, this defaults to 32 KB, and SSDs by default don't defer writes. If write latency is important and your SSD is sufficiently fast, then by increasing this value, you can increase the size of I/Os that you wish to defer to WAL.