From 6892c68bc2b30528dd1200caa9182dcbc4841e71 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Mikl=C3=B3s=20Homolya?= Date: Sat, 19 Sep 2015 11:31:01 +0100 Subject: [PATCH] Update README.md --- README.md | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index 2d7cafc..41c72c0 100644 --- a/README.md +++ b/README.md @@ -4,15 +4,21 @@ #### Usage -Type `make` to generate the meshes for the experiments. Gmsh is required for this step. + make + +will generate the meshes for the experiments. Gmsh is required for this step. `submit.pbs.in` contains a submission template for PBS. To submit an experiment, type ./submit.sh n -where `n` denotes the number of computing nodes. The number of cores will be 24 * `n`. +where `n` denotes the number of computing nodes. The number of cores will be 24 * `n`. The submission will run `measure.py` which facilitates the measurement. It also relies on Firedrake having patched with `firedrake.patch`, which adds some instrumentation code. + +On ARCHER, the above will generate files `measure_`_nnnn_`.o`_kkkkkk_ and `measure_`_nnnn_`.e`_kkkkkk_, containing the standard output and standard error of the job. Our results are uploaded into the results folder. Running + + python evaluate.py -On ARCHER, the above will generate files `measure_`_nnnn_`.o`_kkkkkk_ and `measure_`_nnnn_`.e`_kkkkkk_, containing the standard output and standard error of the job. Our results are uploaded into the results folder. Running `python evaluate.py` will look for these files, process them, show some diagnostic output, and generate the CSV files which are directly used for plots and tables in the paper. +will look for these files, process them, show some diagnostic output, and generate the CSV files which are directly used for plots and tables in the paper. #### Licence