Depending on the origin of the data, several scenarios are possible:
Your data is on the Internet.
There are online data catalogs that allow you to load data lazily (i.e. transparently when the process is running):
In this case, point to the dataset in the notebook as specified on the catalog. For example:
import xarray as xr
store = 'gs://leap-persistent/data-library/feedstocks/eNATL_feedstock/eNATL60-BLBT02.zarr'
ds = xr.open_dataset(store, engine='zarr', chunks={})This way, all you have to do is manipulate the ds variable afterwards.
If the data is on a remote server, then after configuring your SSH keys on your OpenReproLab home, run a Terminal and use the SCP or RSYNC commands:
cd
rsync <need example here>
# or
scp <need example here>Launch these download commands in the right folder on the platform, remembering that your home is only 50GB in size.
Be sure to be in the right sub-directory in the explorer part of the cloud.
Be sure to be in the right sub-directory in the explorer part of the cloud.
Example :
wget https://ige-meom-opendap.univ-grenoble-alpes.fr/thredds/fileServer/meomopendap/extract/MEOM/eNATL60/eNATL60-BLBT02/1h/SICILe/eNATL60SICILe-BLBT02_y2009m08d29.1h_SSH.ncIGE-calcul1 is a cluster accessible to anyone in the IGE
If you are outside the lab or via WIFI, you need to connect first to ige-ssh: ssh agalan@ige-ssh.u-ga.fr
Connect and configure your access to ige-calcul1 following the first steps of this tutorial
A mirror of your private-storage that you have in the OpenReproLab cloud is located at /mnt/summer/openreprolab/USER_GITHUB on calcul1
Then you can upload data from your personnal computer with : rsync -rav MONDOSSIER calcul1:/mnt/summer/openreprolab/USER_GITHUB/ or from another place on calcul1 : cp data /mnt/summer/openreprolab/USER_GITHUB/.
Be aware that you have different identities on the cloud and on calcul1 :
- on the cloud you are logued via your github account
- on ige-calcul1 you are logued via your agalan account
That means that when you create a sub-directory in your private-storage on the cloud, you cannot write to it on calcul1 and conversly
We suggest that you create two sub-directories :
- one from calcul1 : INPUTS-calcul1
- one from the cloud : OUTPUTS-cloud
This way, in a notebook you will be able to read the files in INPUTS and write to OUTPUTS (choose the right names for your directories !)