Latency is the second performance characteristic of a storage device.
It's defined as the time needed to access a particular location in storage.
The relevant unit of measurement is typically:
- nanosecond for primary storage (memory/RAM),
- millisecond for secondary storage (hard disk),
- and second for tertiary storage (removable media).
It may make sense to separate:
- read latency
- and write latency,
and in case of sequential access storage:
- and average latency.
Time scale of system latencies
from Systems Performance: Enterprise and the Cloud (Brendan Gregg)