%matplotlib inline
import pandas as pd
import socket
host = socket.getfqdn()
from core import load, zoom, calc, save,plots,monitor
#reload funcs after updating ./core/*.py
import importlib
importlib.reload(load)
importlib.reload(zoom)
importlib.reload(calc)
importlib.reload(save)
importlib.reload(plots)
importlib.reload(monitor)
<module 'core.monitor' from '/ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/monitor.py'>
If you submit the job with job scheduler; below are list of enviroment variable one can pass
local : if True run dask local cluster, if not true, put number of workers setted in the 'local' if no 'local ' given, local will be setted automatically to 'True'
%env ychunk='2', #%env tchunk='2'
controls chunk. 'False' sets no modification from original netcdf file's chunk.
ychunk=10 will group the original netcdf file to 10 by 10
tchunk=1 will chunk the time coordinate one by one
%env file_exp=
'file_exp': Which 'experiment' name is it? this corresopnds to intake catalog name without path and .yaml
#%env year=
for Validation, this correspoinds to path/year/month 's year for monitoring, this corresponids to 'date' having means do all files in the monitoring directory setting it as 0[0-9] &1[0-9]& [2-3][0-9], the job can be separated in three lots. For DELTA experiment, year corresponds to really 'year'
%env month=
for monitoring this corresponds to file path path-XIOS.{month}/
For DELTA experiment, year corresponds to really 'month'
proceed saving? True or False , Default is setted as True
proceed plotting? True or False , Default is setted as True
proceed computation? or just load computed result? True or False , Default is setted as True
save output file used for plotting
using kerchunked file -> False, not using kerhcunk -> True
name of control file to be used for computation/plots/save/ We have number of M_xxx.csv
Monitor.sh calls M_MLD_2D
and AWTD.sh, Fluxnet.sh, Siconc.sh, IceClim.sh, FWC_SSH.sh, Integrals.sh , Sections.sh
M_AWTMD
M_Fluxnet
M_Ice_quantities
M_IceClim M_IceConce M_IceThick
M_FWC_2D M_FWC_integrals M_FWC_SSH M_SSH_anomaly
M_Mean_temp_velo M_Mooring
M_Sectionx M_Sectiony
%%time
# 'savefig': Do we save output in html? or not. keep it true.
savefig=True
client,cluster,control,catalog_url,month,year,daskreport,outputpath = load.set_control(host)
!mkdir -p $outputpath
!mkdir -p $daskreport
client
local True using host= irene4750.c-irene.mg1.tgcc.ccc.cea.fr starting dask cluster on local= True workers 16 10000000000 rome local cluster starting This code is running on irene4750.c-irene.mg1.tgcc.ccc.cea.fr using SEDNA_DELTA_MONITOR file experiment, read from ../lib/SEDNA_DELTA_MONITOR.yaml on year= * on month= 04 outputpath= ../results/SEDNA_DELTA_MONITOR/ daskreport= ../results/dask/6461355irene4750.c-irene.mg1.tgcc.ccc.cea.fr_SEDNA_DELTA_MONITOR_04M_FWC_integrals/ CPU times: user 552 ms, sys: 133 ms, total: 685 ms Wall time: 21.5 s
Client-460dddf8-17f3-11ed-b2b4-080038b99565
Connection method: Cluster object | Cluster type: distributed.LocalCluster |
Dashboard: http://127.0.0.1:8787/status |
89e6c242
Dashboard: http://127.0.0.1:8787/status | Workers: 16 |
Total threads: 128 | Total memory: 251.06 GiB |
Status: running | Using processes: True |
Scheduler-c8ee2c3f-186b-4540-8b29-78da3f014201
Comm: tcp://127.0.0.1:36101 | Workers: 16 |
Dashboard: http://127.0.0.1:8787/status | Total threads: 128 |
Started: Just now | Total memory: 251.06 GiB |
Comm: tcp://127.0.0.1:34519 | Total threads: 8 |
Dashboard: http://127.0.0.1:39458/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:45387 | |
Local directory: /tmp/dask-worker-space/worker-ac_q_dsp |
Comm: tcp://127.0.0.1:34128 | Total threads: 8 |
Dashboard: http://127.0.0.1:42700/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:36136 | |
Local directory: /tmp/dask-worker-space/worker-0luqjjya |
Comm: tcp://127.0.0.1:36278 | Total threads: 8 |
Dashboard: http://127.0.0.1:41795/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:33730 | |
Local directory: /tmp/dask-worker-space/worker-erzm_jgm |
Comm: tcp://127.0.0.1:43031 | Total threads: 8 |
Dashboard: http://127.0.0.1:40113/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:39141 | |
Local directory: /tmp/dask-worker-space/worker-aeiw4njy |
Comm: tcp://127.0.0.1:45725 | Total threads: 8 |
Dashboard: http://127.0.0.1:37302/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:40152 | |
Local directory: /tmp/dask-worker-space/worker-g68725ss |
Comm: tcp://127.0.0.1:40501 | Total threads: 8 |
Dashboard: http://127.0.0.1:39456/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:43125 | |
Local directory: /tmp/dask-worker-space/worker-nlk5oxhz |
Comm: tcp://127.0.0.1:41404 | Total threads: 8 |
Dashboard: http://127.0.0.1:46754/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:39863 | |
Local directory: /tmp/dask-worker-space/worker-dhuxetrw |
Comm: tcp://127.0.0.1:37283 | Total threads: 8 |
Dashboard: http://127.0.0.1:40799/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:34094 | |
Local directory: /tmp/dask-worker-space/worker-sgtw0hu8 |
Comm: tcp://127.0.0.1:33992 | Total threads: 8 |
Dashboard: http://127.0.0.1:43558/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:43900 | |
Local directory: /tmp/dask-worker-space/worker-49w1q9ax |
Comm: tcp://127.0.0.1:41917 | Total threads: 8 |
Dashboard: http://127.0.0.1:41716/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:34201 | |
Local directory: /tmp/dask-worker-space/worker-oodj3sld |
Comm: tcp://127.0.0.1:46404 | Total threads: 8 |
Dashboard: http://127.0.0.1:33747/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:34527 | |
Local directory: /tmp/dask-worker-space/worker-1d169cto |
Comm: tcp://127.0.0.1:41827 | Total threads: 8 |
Dashboard: http://127.0.0.1:42472/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:41532 | |
Local directory: /tmp/dask-worker-space/worker-tigfmn5t |
Comm: tcp://127.0.0.1:42427 | Total threads: 8 |
Dashboard: http://127.0.0.1:41525/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:46525 | |
Local directory: /tmp/dask-worker-space/worker-s4djmcnh |
Comm: tcp://127.0.0.1:44643 | Total threads: 8 |
Dashboard: http://127.0.0.1:41724/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:37896 | |
Local directory: /tmp/dask-worker-space/worker-hzhw7miz |
Comm: tcp://127.0.0.1:43524 | Total threads: 8 |
Dashboard: http://127.0.0.1:34771/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:40513 | |
Local directory: /tmp/dask-worker-space/worker-o2npjdca |
Comm: tcp://127.0.0.1:35174 | Total threads: 8 |
Dashboard: http://127.0.0.1:34664/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:38147 | |
Local directory: /tmp/dask-worker-space/worker-ewwartht |
df=load.controlfile(control)
#Take out 'later' tagged computations
#df=df[~df['Value'].str.contains('later')]
df
Value | Inputs | Equation | Zone | Plot | Colourmap | MinMax | Unit | Oldname | Unnamed: 10 | |
---|---|---|---|---|---|---|---|---|---|---|
FWC_integrals | calc.FWC_load_integrals(data,nc_outputpath) | BBFG | FWC_integrals | (12000,24000) | Km^3 | I-1 |
Each computation consists of
%%time
import os
calcswitch=os.environ.get('calc', 'True')
lazy=os.environ.get('lazy','False' )
loaddata=((df.Inputs != '').any())
print('calcswitch=',calcswitch,'df.Inputs != nothing',loaddata, 'lazy=',lazy)
data = load.datas(catalog_url,df.Inputs,month,year,daskreport,lazy=lazy) if ((calcswitch=='True' )*loaddata) else 0
data
calcswitch= False df.Inputs != nothing False lazy= False CPU times: user 444 µs, sys: 0 ns, total: 444 µs Wall time: 443 µs
0
%%time
monitor.auto(df,data,savefig,daskreport,outputpath,file_exp='SEDNA'
)
#calc= False #save= False #plot= True Value='FWC_integrals' Zone='BBFG' Plot='FWC_integrals' cmap='' clabel='Km^3' clim= (12000, 24000) outputpath='../results/SEDNA_DELTA_MONITOR/' nc_outputpath='../nc_results/SEDNA_DELTA_MONITOR/' filename='SEDNA_FWC_integrals_BBFG_FWC_integrals' #3 no computing , loading starts data=save.load_data(plot=Plot,path=nc_outputpath,filename=filename) start saving data load 1Dnc file from ../nc_results/SEDNA_DELTA_MONITOR/../*/SEDNA_FWC_integrals_BBFG_FWC_integrals*.nc
--------------------------------------------------------------------------- OSError Traceback (most recent call last) File <timed eval>:1, in <module> File /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/monitor.py:79, in auto(df, val, savefig, daskreport, outputpath, file_exp) 77 print('data=save.load_data(plot=Plot,path=nc_outputpath,filename=filename)' ) 78 with performance_report(filename=daskreport+"_calc_"+step.Value+".html"): ---> 79 data=save.load_data(plot=Plot,path=nc_outputpath,filename=filename) 80 #saveswitch=False 82 display(data) File /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/save.py:32, in load_data(plot, path, filename) 30 print('start saving data') 31 if 'int' in plot: ---> 32 data=load_integral(path,filename) 33 elif 'Mooring' in plot: 34 data=load_integral(path,filename) File /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/save.py:74, in load_integral(path, filename) 72 filesave=path+'../*/'+filename+'*.nc' 73 print('load 1Dnc file from',filesave) ---> 74 return xr.open_mfdataset(filesave 75 ,compat='override' 76 ,data_vars='minimal' 77 ,concat_dim=('t') 78 ,combine='nested' 79 ,coords='minimal') File /ccc/cont003/home/ra5563/ra5563/monitor/lib/python3.10/site-packages/xarray/backends/api.py:937, in open_mfdataset(paths, chunks, concat_dim, compat, preprocess, engine, data_vars, coords, combine, parallel, join, attrs_file, combine_attrs, **kwargs) 934 paths = [os.fspath(p) if isinstance(p, os.PathLike) else p for p in paths] 936 if not paths: --> 937 raise OSError("no files to open") 939 if combine == "nested": 940 if isinstance(concat_dim, (str, DataArray)) or concat_dim is None: OSError: no files to open