Incomputing,elapsed real time,real time,wall-clock time,wall time, orwalltime is the actual time taken from the start of acomputer program to the end. In other words, it is the difference between the time at which a task finishes and the time at which the task started.
Wall time is thus different fromCPU time, which measures only the time during which the processor is actively working on a certain task orprocess. The difference between the two can arise fromarchitecture andrun-time dependent factors, e.g.programmed delays or waiting forsystem resources to become available. Consider the example of a mathematical program that reports that it has used "CPU time 0m0.04s, Wall time 6m6.01s". This means that while the program was active for six minutes and six point zero, one second, during that time the computer's processor spent only 4/100 of a second performing calculations for the program.[citation needed]
Conversely, programsrunning in parallel on more than one processing unit can spend CPU time many times beyond their elapsed time. Since inconcurrent computing the definition of elapsed time is non-trivial, the conceptualization of the elapsed time as measured on a separate, independent wall clock is convenient.
Another definition of "wall time" is the measurement of time via a separate, independent clock as opposed to the localsystem time (internal), i.e. with regard to the difference between the two.[1]
The term wall-clock time has also found widespread adoption in computer simulation, to distinguish between (1) the (often compressed or expanded) simulation time, and (2) the time as it passes for the user of the simulation tool.[2][3][4]
 | This computing article is astub. You can help Wikipedia byadding missing information. |
