Editing Latency (engineering) (section)

==Computer hardware and software systems==
{{Further|Access time}}

Computers run [[Machine code|instructions]] in the context of a [[Process (computing)|process]].  In the context of [[computer multitasking]], the execution of the process can be postponed if other processes are also executing.  In addition, the operating system can schedule when to perform the action that the process is commanding.  For example, suppose a process commands that a computer card's voltage output be set high-low-high-low and so on at a rate of 1000&nbsp;Hz.  The operating system [[Scheduling (computing)|schedules]] the process for each transition (high-low or low-high) based on a hardware clock such as the [[High Precision Event Timer]].  The latency is the delay between the events generated by the hardware clock and the actual transitions of voltage from high to low or low to high.

Many [[desktop operating system]]s have performance limitations that create additional latency. The problem may be mitigated with real-time extensions and patches such as [[PREEMPT RT]].

On embedded systems, the real-time execution of instructions is often supported by a [[real-time operating system]].

Note that in [[software system]]s, benchmarking against "average" and "median" latency can be misleading because few outlier numbers can distort them. Instead, [[software architect]]s and [[software developer]]s should use "99th percentile".<ref>{{Cite book |title=Foundations of Data Intensive Applications Large Scale Data Analytics Under the Hood |year=2021 |isbn=9781119713012}}</ref>