%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'>
# 'month': = 'JOBID' almost month but not really,
# If you submit the job with job scheduler, above
#below are list of enviroment variable one can pass
#%env local='2"
# 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 control=FWC_SSH
# name of control file to be used for computation/plots/save/
#%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.
#%env month=
# for monitoring this corresponds to file path path-XIOS.{month}/
#
#%env save= proceed saving? True or False , Default is setted as True
#%env plot= proceed plotting? True or False , Default is setted as True
#%env calc= proceed computation? or just load computed result? True or False , Default is setted as True
#%env save=False
#%env lazy=False
%%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= irene4142.c-irene.mg1.tgcc.ccc.cea.fr starting dask cluster on local= True workers 16 10000000000 False rome local cluster starting This code is running on irene4142.c-irene.mg1.tgcc.ccc.cea.fr using SEDNA_DELTA_MONITOR file experiment, read from ../lib/SEDNA_DELTA_MONITOR.yaml on year= 2012 on month= 02 outputpath= ../results/SEDNA_DELTA_MONITOR/ daskreport= ../results/dask/6414597irene4142.c-irene.mg1.tgcc.ccc.cea.fr_SEDNA_DELTA_MONITOR_02M_AWTMD/ CPU times: user 531 ms, sys: 117 ms, total: 648 ms Wall time: 20.9 s
Client-3a8d5a2e-1358-11ed-915f-080038b9322d
Connection method: Cluster object | Cluster type: distributed.LocalCluster |
Dashboard: http://127.0.0.1:8787/status |
b55104e4
Dashboard: http://127.0.0.1:8787/status | Workers: 16 |
Total threads: 128 | Total memory: 251.06 GiB |
Status: running | Using processes: True |
Scheduler-c971a72b-8431-452b-90e0-01317736ca3d
Comm: tcp://127.0.0.1:41769 | Workers: 16 |
Dashboard: http://127.0.0.1:8787/status | Total threads: 128 |
Started: Just now | Total memory: 251.06 GiB |
Comm: tcp://127.0.0.1:44814 | Total threads: 8 |
Dashboard: http://127.0.0.1:44486/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:43510 | |
Local directory: /tmp/dask-worker-space/worker-jfvkywpp |
Comm: tcp://127.0.0.1:43178 | Total threads: 8 |
Dashboard: http://127.0.0.1:42874/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:36238 | |
Local directory: /tmp/dask-worker-space/worker-8nevk2c4 |
Comm: tcp://127.0.0.1:42894 | Total threads: 8 |
Dashboard: http://127.0.0.1:43610/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:33854 | |
Local directory: /tmp/dask-worker-space/worker-a9lk5h6h |
Comm: tcp://127.0.0.1:32920 | Total threads: 8 |
Dashboard: http://127.0.0.1:46862/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:41216 | |
Local directory: /tmp/dask-worker-space/worker-64g74q28 |
Comm: tcp://127.0.0.1:46002 | Total threads: 8 |
Dashboard: http://127.0.0.1:38051/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:42895 | |
Local directory: /tmp/dask-worker-space/worker-d6bj2jka |
Comm: tcp://127.0.0.1:40274 | Total threads: 8 |
Dashboard: http://127.0.0.1:43067/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:36750 | |
Local directory: /tmp/dask-worker-space/worker-u4hkwcln |
Comm: tcp://127.0.0.1:36161 | Total threads: 8 |
Dashboard: http://127.0.0.1:45874/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:42715 | |
Local directory: /tmp/dask-worker-space/worker-1wfbbhu5 |
Comm: tcp://127.0.0.1:35999 | Total threads: 8 |
Dashboard: http://127.0.0.1:43481/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:44571 | |
Local directory: /tmp/dask-worker-space/worker-ap4a6n4n |
Comm: tcp://127.0.0.1:45217 | Total threads: 8 |
Dashboard: http://127.0.0.1:36561/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:39692 | |
Local directory: /tmp/dask-worker-space/worker-syb7v92w |
Comm: tcp://127.0.0.1:43873 | Total threads: 8 |
Dashboard: http://127.0.0.1:43813/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:43388 | |
Local directory: /tmp/dask-worker-space/worker-qk7qixct |
Comm: tcp://127.0.0.1:37401 | Total threads: 8 |
Dashboard: http://127.0.0.1:46242/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:40938 | |
Local directory: /tmp/dask-worker-space/worker-ghw_m3uj |
Comm: tcp://127.0.0.1:42759 | Total threads: 8 |
Dashboard: http://127.0.0.1:41320/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:32931 | |
Local directory: /tmp/dask-worker-space/worker-6tdmp2qj |
Comm: tcp://127.0.0.1:35363 | Total threads: 8 |
Dashboard: http://127.0.0.1:34578/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:44489 | |
Local directory: /tmp/dask-worker-space/worker-ovxpsv54 |
Comm: tcp://127.0.0.1:40104 | Total threads: 8 |
Dashboard: http://127.0.0.1:43837/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:43443 | |
Local directory: /tmp/dask-worker-space/worker-d3dlu4tk |
Comm: tcp://127.0.0.1:32928 | Total threads: 8 |
Dashboard: http://127.0.0.1:39092/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:43093 | |
Local directory: /tmp/dask-worker-space/worker-rarrk8fg |
Comm: tcp://127.0.0.1:39966 | Total threads: 8 |
Dashboard: http://127.0.0.1:40043/status | Memory: 15.69 GiB |
Nanny: tcp://127.0.0.1:33119 | |
Local directory: /tmp/dask-worker-space/worker-54qsjqcu |
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 | |
---|---|---|---|---|---|---|---|---|---|---|
AW_maxtemp_depth | gridT.votemper,gridS.vosaline,param.mask,param... | calc.AWTD4(data) | ALL | AWTD_map | jet | (0,800) | m | M-5 |
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 True lazy= False CPU times: user 299 µs, sys: 48 µs, total: 347 µs Wall time: 347 µs
0
%%time
monitor.auto(df,data,savefig,daskreport,outputpath,file_exp='SEDNA'
)
#calc= False #save= False #plot= True Value='AW_maxtemp_depth' Zone='ALL' Plot='AWTD_map' cmap='jet' clabel='m' clim= (0, 800) outputpath='../results/SEDNA_DELTA_MONITOR/' nc_outputpath='../nc_results/SEDNA_DELTA_MONITOR/' filename='SEDNA_AWTD_map_ALL_AW_maxtemp_depth' #3 no computing , loading starts dtaa=save.load_data(plot=Plot,path=nc_outputpath,filename=filename) start saving data filename= ../nc_results/SEDNA_DELTA_MONITOR/SEDNA_AWTD_map_ALL_AW_maxtemp_depth/t_*/y_*/x_*.nc
--------------------------------------------------------------------------- ValueError Traceback (most recent call last) File <timed eval>:1, in <module> File /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/monitor.py:76, in auto(df, val, savefig, daskreport, outputpath, file_exp) 74 print('dtaa=save.load_data(plot=Plot,path=nc_outputpath,filename=filename)' ) 75 with performance_report(filename=daskreport+"_calc_"+step.Value+".html"): ---> 76 data=save.load_data(plot=Plot,path=nc_outputpath,filename=filename) 77 #saveswitch=False 79 display(data) File /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/save.py:38, in load_data(plot, path, filename) 36 data=load_twoD(path,filename,nested=False) 37 else: ---> 38 data=load_twoD(path,filename) 39 print('load computed data completed') 40 return data File /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/save.py:47, in load_twoD(path, filename, nested) 45 filename=filesave+'/t_*/y_*/x_*.nc' if nested else filesave+'/t_*.nc' 46 print ('filename=',filename) ---> 47 return xr.open_mfdataset(filename,parallel=True 48 ,compat='override' 49 ,data_vars='minimal' 50 ,concat_dim=('x','y','t') 51 ,combine='nested' #param_xios 52 ,coords='minimal') File /ccc/cont003/home/ra5563/ra5563/monitor/lib/python3.10/site-packages/xarray/backends/api.py:987, in open_mfdataset(paths, chunks, concat_dim, compat, preprocess, engine, data_vars, coords, combine, parallel, join, attrs_file, combine_attrs, **kwargs) 983 try: 984 if combine == "nested": 985 # Combined nested list by successive concat and merge operations 986 # along each dimension, using structure given by "ids" --> 987 combined = _nested_combine( 988 datasets, 989 concat_dims=concat_dim, 990 compat=compat, 991 data_vars=data_vars, 992 coords=coords, 993 ids=ids, 994 join=join, 995 combine_attrs=combine_attrs, 996 ) 997 elif combine == "by_coords": 998 # Redo ordering from coordinates, ignoring how they were ordered 999 # previously 1000 combined = combine_by_coords( 1001 datasets, 1002 compat=compat, (...) 1006 combine_attrs=combine_attrs, 1007 ) File /ccc/cont003/home/ra5563/ra5563/monitor/lib/python3.10/site-packages/xarray/core/combine.py:365, in _nested_combine(datasets, concat_dims, compat, data_vars, coords, ids, fill_value, join, combine_attrs) 362 _check_shape_tile_ids(combined_ids) 364 # Apply series of concatenate or merge operations along each dimension --> 365 combined = _combine_nd( 366 combined_ids, 367 concat_dims, 368 compat=compat, 369 data_vars=data_vars, 370 coords=coords, 371 fill_value=fill_value, 372 join=join, 373 combine_attrs=combine_attrs, 374 ) 375 return combined File /ccc/cont003/home/ra5563/ra5563/monitor/lib/python3.10/site-packages/xarray/core/combine.py:228, in _combine_nd(combined_ids, concat_dims, data_vars, coords, compat, fill_value, join, combine_attrs) 226 n_dims = len(example_tile_id) 227 if len(concat_dims) != n_dims: --> 228 raise ValueError( 229 "concat_dims has length {} but the datasets " 230 "passed are nested in a {}-dimensional structure".format( 231 len(concat_dims), n_dims 232 ) 233 ) 235 # Each iteration of this loop reduces the length of the tile_ids tuples 236 # by one. It always combines along the first dimension, removing the first 237 # element of the tuple 238 for concat_dim in concat_dims: ValueError: concat_dims has length 3 but the datasets passed are nested in a 1-dimensional structure