%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= irene4453.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 irene4453.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/6413749irene4453.c-irene.mg1.tgcc.ccc.cea.fr_SEDNA_DELTA_MONITOR_02M_IceClim/ CPU times: user 507 ms, sys: 135 ms, total: 643 ms Wall time: 19.7 s
Client-d643c42a-1344-11ed-b441-080038b93255
| Connection method: Cluster object | Cluster type: distributed.LocalCluster |
| Dashboard: http://127.0.0.1:8787/status |
42068dea
| Dashboard: http://127.0.0.1:8787/status | Workers: 16 |
| Total threads: 128 | Total memory: 251.06 GiB |
| Status: running | Using processes: True |
Scheduler-a271ac39-f968-4d0d-b96b-c61c43c0bb33
| Comm: tcp://127.0.0.1:41680 | Workers: 16 |
| Dashboard: http://127.0.0.1:8787/status | Total threads: 128 |
| Started: Just now | Total memory: 251.06 GiB |
| Comm: tcp://127.0.0.1:39936 | Total threads: 8 |
| Dashboard: http://127.0.0.1:40286/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:42339 | |
| Local directory: /tmp/dask-worker-space/worker-mawfo74c | |
| Comm: tcp://127.0.0.1:46577 | Total threads: 8 |
| Dashboard: http://127.0.0.1:42365/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:41463 | |
| Local directory: /tmp/dask-worker-space/worker-kbpzle9a | |
| Comm: tcp://127.0.0.1:36164 | Total threads: 8 |
| Dashboard: http://127.0.0.1:36032/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:41302 | |
| Local directory: /tmp/dask-worker-space/worker-aa8ivemh | |
| Comm: tcp://127.0.0.1:37347 | Total threads: 8 |
| Dashboard: http://127.0.0.1:45755/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:34136 | |
| Local directory: /tmp/dask-worker-space/worker-usupdfct | |
| Comm: tcp://127.0.0.1:33328 | Total threads: 8 |
| Dashboard: http://127.0.0.1:36835/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:42357 | |
| Local directory: /tmp/dask-worker-space/worker-j5vccs6b | |
| Comm: tcp://127.0.0.1:38047 | Total threads: 8 |
| Dashboard: http://127.0.0.1:34423/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:46394 | |
| Local directory: /tmp/dask-worker-space/worker-zgre43_z | |
| Comm: tcp://127.0.0.1:46116 | Total threads: 8 |
| Dashboard: http://127.0.0.1:36932/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:37000 | |
| Local directory: /tmp/dask-worker-space/worker-2046n_p5 | |
| Comm: tcp://127.0.0.1:33842 | Total threads: 8 |
| Dashboard: http://127.0.0.1:33930/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:33184 | |
| Local directory: /tmp/dask-worker-space/worker-noxysrca | |
| Comm: tcp://127.0.0.1:44088 | Total threads: 8 |
| Dashboard: http://127.0.0.1:33825/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:35108 | |
| Local directory: /tmp/dask-worker-space/worker-09ichjvk | |
| Comm: tcp://127.0.0.1:40163 | Total threads: 8 |
| Dashboard: http://127.0.0.1:40558/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:42039 | |
| Local directory: /tmp/dask-worker-space/worker-1lnas8xd | |
| Comm: tcp://127.0.0.1:38751 | Total threads: 8 |
| Dashboard: http://127.0.0.1:44581/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:36040 | |
| Local directory: /tmp/dask-worker-space/worker-ev4kw286 | |
| Comm: tcp://127.0.0.1:37363 | Total threads: 8 |
| Dashboard: http://127.0.0.1:39367/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:34949 | |
| Local directory: /tmp/dask-worker-space/worker-vijx23op | |
| Comm: tcp://127.0.0.1:35819 | Total threads: 8 |
| Dashboard: http://127.0.0.1:40755/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:42759 | |
| Local directory: /tmp/dask-worker-space/worker-mdc10pve | |
| Comm: tcp://127.0.0.1:36868 | Total threads: 8 |
| Dashboard: http://127.0.0.1:46211/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:36332 | |
| Local directory: /tmp/dask-worker-space/worker-0th1pumi | |
| Comm: tcp://127.0.0.1:35408 | Total threads: 8 |
| Dashboard: http://127.0.0.1:39453/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:34425 | |
| Local directory: /tmp/dask-worker-space/worker-rmgzltvp | |
| Comm: tcp://127.0.0.1:45709 | Total threads: 8 |
| Dashboard: http://127.0.0.1:44389/status | Memory: 15.69 GiB |
| Nanny: tcp://127.0.0.1:34595 | |
| Local directory: /tmp/dask-worker-space/worker-ina1jkri | |
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 | |
|---|---|---|---|---|---|---|---|---|---|---|
| IceClim | calc.IceClim_load(data,nc_outputpath) | ALL | IceClim | Spectral | (0,5) | m | M-4 |
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= True df.Inputs != nothing False lazy= False CPU times: user 0 ns, sys: 481 µs, total: 481 µs Wall time: 479 µs
0
%%time
monitor.auto(df,data,savefig,daskreport,outputpath,file_exp='SEDNA'
)
#calc= True #save= False #plot= True
--------------------------------------------------------------------------- AttributeError Traceback (most recent call last) File <timed eval>:1, in <module> File /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/monitor.py:26, in auto(df, val, savefig, daskreport, outputpath, file_exp) 23 print('#plot=',plotswitch ) 24 for step in df.itertuples(): 25 # 1. Create data set ---> 26 optimize_dataset(val) 27 data=val 28 Value=step.Value File /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/monitor.py:100, in optimize_dataset(ds) 98 def optimize_dataset(ds): 99 import dask --> 100 for varname, da in ds.data_vars.items(): 101 #print(varname) 102 da=da.data 103 (da,)=dask.optimize(da) AttributeError: 'int' object has no attribute 'data_vars'