www.ti.com
Execution Time
Execution time should be expressed in instruction cycles whereas the period expressed in microseconds. Worst-case execution time must be accompanied with a precise description of the run-time assumptions required to reproduce this upper bound. For example, placement of code and data in internal or external memory, placement of specified buffers in dual-access or single access on-chip memory, etc. In particular, the worst-case execution time must be accompanied by a table of memory requirements (described above) necessary to achieve the quoted execution time. Note that the entries in this table are not required to be constants; they may be functions of the algorithm'sinstance creation parameters.
Operation | Period | Worst-Case Cycles/Period |
process() | 22500 μs | 198000 |
In some cases, an algorithm'sworst-case execution time is a periodic function of the frame number. Suppose, for example, that an audio encoder consumes 10 milliseconds frames of data at a time but only outputs encoded data on every 20 milliseconds. In this case, the encoder'sworst-case execution time on even frames will differ (perhaps significantly) from the worst-case execution time for odd numbered frames; the output of data only occurs on odd frames. In these situations, it is important to characterize the worst-case execution time for each frame; otherwise, system integrators may (falsely) conclude that an algorithm will not be able to be combined with others.
All such algorithms must characterize their periodic execution time requirements by filling in the table below; the number of Cycles/Period columns can be any finite number M. The worst-case number in the Cycles/PeriodN column must be the worst-case number of cycles that can occur on frame number k * M + N, where k is any positive integer.
Operation | Period | Cycles/Period0 | Cycles/Period1 |
process() | 22500 μs | 59000 | 198000 |
SPRU352G –June 2005 –Revised February 2007 | Algorithm Performance Characterization | 43 |
| | |