Difference between revisions of "Talk:Real Time Terms"

From eLinux.org
Jump to: navigation, search
(make note on time)
 
 
Line 1: Line 1:
 
These terms do not state a common model for time measurements.  I believe it is relatively well-established
 
These terms do not state a common model for time measurements.  I believe it is relatively well-established
 
that the following events and time periods are of interest when measuring realtime performance:
 
that the following events and time periods are of interest when measuring realtime performance:
 
== Events ==
 
* A - hardware interrupt assertion
 
** This is the time when the hardware interrupt line for an event is raised
 
* B - Interrupt service routine starts execution
 
* C - Process starts execution
 
* D - Result is received from processing the event
 
 
== Time Periods ==
 
Real time performance is often expressed in terms of the maximum, minimum and average duration
 
for certain time periods defined by the events above.
 
 
Here are some common terms, expressed relative to those events:
 
; Interrupt latency : The time from A to B
 
; Scheduling latency : The time from B to C (this is sometimes referred to as the time from A to C)
 
; Processing time : The time from C to D
 
; Response latency : The time from A to D
 
 
Often, these or similar terms are used with less accuracy to describe the closest
 
approximation one can get to these, with a particular test framework and instrumentation set.
 
 
For example, Response latency may be reported as "response time", and refer to the time from
 
when a host program records the time, previous to transmitting a piece of data which will cause
 
an interrupt on the target machine, to the time when a host program records the time after
 
receiving some signal from the target that processing is completed.  Of course, time is taken
 
in the operations of recording time, tranmissing data, and detecting signals on the host machine.
 
But it may be that timing these individual operations at a finer granularity requires
 
instrumentation or hardware support not reasonably available for a test.
 

Latest revision as of 18:33, 8 June 2007

These terms do not state a common model for time measurements. I believe it is relatively well-established that the following events and time periods are of interest when measuring realtime performance: