Exercise - Puhti web interface#

Timing

Goals

Getting familiar with the Puhti web interface
First contact with the supercomputer
- Files: moving, viewing, editing
- Graphical tools: Visual Studio Code, QGIS, Jupyter, RStudio
- See info about Puhti and your project

Prerequisites

Get familiar with Puhti web interface#

Change the default project and username

project_200xxxx is example project name, replace with your own CSC project name.
cscusername is example username, replace with your username.

Puhti general status: bottom of front page
- Sometimes when the Disk lag here is high, reading and writing files might get slow.
Own projects, remaining billing units: Tools -> Project view
Disk usage of own projects: Tools -> Disk quotas
Running jobs: Jobs -> Active jobs

Open home directory: Files -> Home Directory
Create new myfile.txt file and add some text to it.
Create new directory mydata
Move the new file under mydata:
- Mark check-box in front of the file
- Click Copy/Move
- Open mydata
- Click Move
Open your mydata folder
Download your file to your local computer

Moving data

Web interface is for moving up to 10Gb data, if you have more data use other tools. More info in moving data

For Python: any CSC module or own installation
CSC Docs: Jupyter
Open the Jupyter launch page: from front page or Apps -> Jupyter
Use settings:
- (Reservation: geocomputing_thu, only during course)
- Project: project_200xxxx
- Partition: interactive (small during course)
- Number of CPU cores: 1
- Memory (Gb): 2
- Local disk: 0
- Time: 0:15:00
- Python: geoconda
- Jupyter type: Lab
- Working directory: /users/cscusername
- Launch
Wait a moment for Jupyter to start -> Connect to Jupyter
Create new Notebook: + -> Notebook: Python 3
Open Statistic Finland Paavo post code data and plot it, add code and run the it with Shift + Enter.

import geopandas
src = geopandas.read_file('/appl/data/geo/tilastokeskus/paavo/2023/pno_tilasto_2023.shp')
src.plot()

End the session:
- Delete the Notebook.
- Close the web tab
- In Active sessions view: Delete

QGIS, GRASS GIS, SAGA GIS, SNAP etc are available via Desktop
CSC Docs: Desktop
Open the Desktop launch page: from front page or Apps -> Desktop
Use settings:
- (Reservation: geocomputing_thu, only during course)
- Project: project_200xxxx
- Partition: interactive (small during course)
- Number of CPU cores: 1
- Memory (Gb): 4
- Local disk: 0
- Time: 0:15:00
- Launch
Wait a moment for Desktop to start -> Launch Desktop
Double-click the QGIS icon
Open Statistic Finland Paavo post code data
- Layer -> Add layer -> Add vector layer
  - Source Type: File
  - Source: /appl/data/geo/tilastokeskus/paavo/2023/pno_tilasto_2023.shp
See file information with GDAL
- Processing -> Toolbox -> GDAL -> Vector miscellanious -> Vector information
- The open dataset is selected by default
- Run
- Note, if interested in moving from graphical QGIS to scripting:
  - The GDAL commandline command is displayed in the lower part of dialog box and log.
  - Advanced menu provides this command also as qgis_processing command and as PyQGIS code.

End the session:
- Close QGIS.
- Close the web tab
- In Active sessions view: Delete
  - This only ends the Desktop session, any files written during the session would be available also afterwards.

QGIS in practice on supercomputer

QGIS is designed for desktop use and it mostly uses only 1 core, so running it on supercomputer is rather slower than on desktop. QGIS is in Puhti and LUMI mainly for easy viewing of input and output data.
With qgis_processing or PyQGIS scripts it is possible to paralellize your data analysis. In general other Python packages are faster, but if you have these scripts already available, they can be used.

Look from CSC Docs how to start SNAP or some other tool of your interest in web interface.

Key points