X-Git-Url: http://info.iut-bm.univ-fcomte.fr/pub/gitweb/simgrid.git/blobdiff_plain/f193fed57024f1b113f3ec07500a76e1ed5b76b1..7115f40452d0798fc920253f50aee208ea0588c2:/src/surf/surf.c diff --git a/src/surf/surf.c b/src/surf/surf.c index f6d9f205be..d995e2038b 100644 --- a/src/surf/surf.c +++ b/src/surf/surf.c @@ -117,16 +117,16 @@ s_surf_model_description_t surf_network_model_description[] = { "Simplistic network model where all communication take a constant time (one second)", surf_network_model_init_Constant}, {"CM02", - "Realistic network model with lmm_solve and no correction factors", + "Legacy network model with lmm_solve (slow-start and bottleneck sharing with small bandwidth poorly modeled).", surf_network_model_init_CM02}, {"LV08", - "Realistic network model with lmm_solve, adequate correction factors (latency*=10.4, bandwidth*=.92, S=8775) and partial invalidation optimization", + "Realistic network model (slow-start modeled by multiplying latency by 10.4, bandwidth by .92 and bottleneck sharing uses a payload of S=8775 for evaluating RTT). Uses partial invalidation optimization of lmm system and a heap for action management by default (option --cfg=network/optim:Lazy)", im_surf_network_model_init_LegrandVelho}, {"LV08_fullupdate", - "Realistic network model wit lmm_solve, adequate correction factors (latency*=10.4, bandwidth*=.92, S=8775) but no further optimization. Should produce the same results as LV08, only slower.", - surf_network_model_init_LegrandVelho}, + "Realistic network model (see LV08) [Deprecated. Use LV08 with option --cfg=network/optim:Full. The Full option updates all actions at every step, which may be useful for debug.]", + im_surf_network_model_init_LegrandVelho}, {"SMPI", - "Realistic network model with lmm_solve and correction factors on three intervals (< 1KiB, < 64 KiB, >= 64 KiB)", + "Realistic network model with lmm_solve specifically tailored for HPC setting (accurate modeling of slow start with correction factors on three intervals (< 1KiB, < 64 KiB, >= 64 KiB))", surf_network_model_init_SMPI}, #ifdef HAVE_GTNETS {"GTNets", @@ -139,32 +139,35 @@ s_surf_model_description_t surf_network_model_description[] = { surf_network_model_init_NS3}, #endif {"Reno", - "Model using lagrange_solve instead of lmm_solve (experts only)", + "Model from Steven H. Low using lagrange_solve instead of lmm_solve (experts only; check the code for more info)", surf_network_model_init_Reno}, {"Reno2", - "Model using lagrange_solve instead of lmm_solve (experts only)", + "Model from Steven H. Low using lagrange_solve instead of lmm_solve (experts only; check the code for more info)", surf_network_model_init_Reno2}, {"Vegas", - "Model using lagrange_solve instead of lmm_solve (experts only)", + "Model from Steven H. Low using lagrange_solve instead of lmm_solve (experts only; check the code for more info)", surf_network_model_init_Vegas}, {NULL, NULL, NULL} /* this array must be NULL terminated */ }; s_surf_model_description_t surf_cpu_model_description[] = { - {"Cas01_fullupdate", "CPU classical model time=size/power", + {"Cas01_fullupdate", "CPU classical model time=size/power (Cas01). [Deprecated. Use Cas01 with option --cfg=cpu/optim:Full. The Full option updates all actions at every step, which may be useful for debug.]", surf_cpu_model_init_Cas01}, {"Cas01", - "Variation of Cas01_fullupdate with partial invalidation optimization of lmm system. Should produce the same values, only faster", - surf_cpu_model_init_Cas01_im}, + "CPU classical model time=size/power. Default version uses partial invalidation optimization of lmm system and a heap for action management (default option is --cfg=cpu/optim:Lazy)", + surf_cpu_model_init_Cas01}, {"CpuTI", - "Variation of Cas01 with also trace integration. Should produce the same values, only faster if you use availability traces", + "CPU classical model time=size/power (Cas01). [Deprecated. Use Cas01 with option --cfg=cpu/optim:TI. The TI optimization is highly optimized when using traces]", surf_cpu_model_init_ti}, {NULL, NULL, NULL} /* this array must be NULL terminated */ }; s_surf_model_description_t surf_workstation_model_description[] = { + {"current_default", + "Curent default workstation model (may change with versions of SimGrid). Currently Cas01 and LV08 (with full duplex support) as CPU and Network", + surf_workstation_model_init_current_default}, {"CLM03", - "Default workstation model, using LV08 and CM02 as network and CPU", + "Default workstation model, using Cas01 and CM02 as CPU and Network", surf_workstation_model_init_CLM03}, {"compound", "Workstation model allowing you to use other network and CPU models", @@ -174,6 +177,19 @@ s_surf_model_description_t surf_workstation_model_description[] = { {NULL, NULL, NULL} /* this array must be NULL terminated */ }; +s_surf_model_description_t surf_optimization_mode_description[] = { + {"Lazy", + "Lazy action management (partial invalidation in lmm + heap in action remaining).", + NULL}, + {"TI", + "Trace integration. Highly optimized mode for the Cas01 model when using availability traces.", + NULL}, + {"Full", + "Full update of remaining and variables. Slow but may be useful when debugging.", + NULL}, + {NULL, NULL, NULL} /* this array must be NULL terminated */ +}; + /** Displays the long description of all registered models, and quit */ void model_help(const char *category, s_surf_model_description_t * table) { @@ -198,7 +214,7 @@ int find_model_description(s_surf_model_description_t * table, for (i = 1; table[i].name; i++) { name_list = xbt_realloc(name_list, - strlen(name_list) + strlen(table[i].name) + 2); + strlen(name_list) + strlen(table[i].name) + 3); strcat(name_list, ", "); strcat(name_list, table[i].name); } @@ -366,6 +382,7 @@ void surf_exit(void) xbt_dynar_foreach(model_list, iter, model) model->model_private->finalize(); xbt_dynar_free(&model_list); + routing_exit(); if (maxmin_system) { lmm_system_free(maxmin_system);