Changes between Version 3 and Version 4 of ticket/0829
- Timestamp:
- 2011-05-31T11:06:15+02:00 (13 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
ticket/0829
v3 v4 22 22 * More radical change : the working space is build dynamically. This solves limitation 2. 23 23 24 ==== Timing ==== 25 ==== Auto-assignement ==== 26 ==== Dynamic dynamic memory ==== 24 ==== 1- Timing ==== 25 26 This functionality doesn't aim to replace advanced software used for optimisation but: 27 * to give a rough idea of performance (CPU and elapsed) 28 * use the same tools and format on all computers( fortran intrinsec CPU_TIME and WMPI_TIME) 29 It is bases on a linked chain of informations to be able to add dynamically new sections and add sub-sections[[BR]] 30 Implementation: 31 * CALL timing_init in nemogcm_init 32 * CALL timing_finalize at the end of nemoggcm 33 * at the end of step, IF( kt == nit000) CALL timing_reset (once the list of varibles has been built) 34 * in each routine to instrument : CALL timing_start('NAME') CALL timing_stop('NAME') 35 Imbricated sub-sections are allowed and their time is then subtracted from the mother section unless the call of timing of the section is done in the following way CALL timing_start('NAME') CALL timing_stop('NAME',section) 36 37 ==== 2- Auto-assignement ==== 38 ==== 3- Dynamic dynamic memory ==== 27 39 28 40