13  Demo Plan for the SRR Meeting

The goal of the SRR meeting is to demonstrate and review the prototype implementation of the APEx services. Below is an outline of the demo scenarios to be presented:

13.1 Demo - APEx Portal

13.1.1 Objective

Showcase the main APEx portal and its core functionality, emphasizing its central role as the gateway to all APEx services.

13.1.2 Scenario

Step Description Partner

13.2 Demo - Project Support through APEx Instantiation Services

13.2.1 Objective

Demonstrate how the instantiation services work and how they provide support from a project’s perspective.

13.2.2 Storyline

A project has recently started and learned that APEx offers instantiation services that could support their activities. The goal of the project is to develop a new algorithm. To achieve this, the project needs a collaborative working environment where they can work together on the algorithm and share results easily. After the approval by their ESA TO, they have contacted the APEx team to request such an environment.

13.2.3 Scenario

Step Description Partner Presenter
1 Instantiate the project environment, including a project portal, user workspace, IDE, and catalogue (API + browser) instance. Inuits Robin
j2 Provide the project team (VITO) with the necessary environment details. Inuits Robin
3 Navigate to the instantiated IDE/Jupyter Notebook environment. VITO
4 Execute a script to generate a COG. VITO
5 Register the results in the instantiated catalogue via a script. VITO
6 Display the result in the STAC browser. VITO
7 Configure and showcase the Geospatial Explorer for visualizing the STAC item/COG. Sparkgeo Adam
8 Launch the Geospatial Explorer and showcase its features. Sparkgeo Adam

13.2.4 Preparation

The following table outlines the prerequisites for preparing the demo:

Action Responsible
Instantiate a testing environment for the demo. Inuits
Prepare the script for generating a COG. VITO
Prepare the script for registering the COG in the catalogue. VITO
Prepare the Geospatial Explorer to support STAC visualization. Sparkgeo

13.3 Demo - Algorithm Propagation

13.3.1 Objective

Showcase how APEx tools and services support algorithm propagation use cases after the project has ended.

Some objectives for the demo:

ID Service Description
UPS1 Upscaling Show that the current Python API is evolving towards a tool can run an openEO UDP over a larger extent, for a fixed tiling grid and a set of fixed parameters. Minimal or no programming should be required.
UPS2 Upscaling Show that a GeoParquet (or STAC API?) based job tracker can enable visual monitoring of a job.
UPS3 Upscaling Show that a processing platform can publish results directly to object storage, allowing other tools (like Geospatial Explorer) to visualize them for quality assurance.
ENH1 Enhancement Show a tool that generates performance metrics for an openEO UDP, allowing to compare performance pre and post optimization, or to check the influence of various job options.
CAT1 Algorithm Catalogue Show that the catalogs checks that the service is available and functioning according to a predefined scenario.
CAT2 Algorithm Catalogue Show that the catalog allows to perform cost checks over time, especially helping to detect unwanted cost increase. (Cost decrease is not a problem.)
CAT3 Algorithm Catalogue Show that the catalog helps users to easily invoke algorithms, without requiring other sources of documentation.
ONB1 Algorithm Onboarding Show the current onboarding process based on openEO UDP.

13.3.2 Storyline

This demo will start with a project interested in using an algorithm from the APEx algorithm catalogue. The demo will highlight the tools available to both APEx administrators and projects for reusing APEx-onboarded algorithms.

13.3.3 Scenario

13.3.3.1 Algorithm catalog, onboarding, enhancement

Presenter:

  1. Go over onboarding of UDP. Clearly indicate that the manual process is not final, but we illustrate that it’s as simple as adding a json in a github repo. Aligns nicely with Earth Code!
  2. To determine cost of UDP, the notebook from ‘enhancement’ is used. Walk through a predefined example.
  3. Browse through the existing algorithms, show catalog UI.
  4. Show algorithm metrics dashboard, covering both cost and availability.
  5. Show how user invokes an algorithm directly, based on info that is readily available in the catalog. Important to stress that the user needs a platform account.

13.3.3.2 Upscaling

Presenter:

  1. Show the preparation phase for upscaling (tiling, dataframe creation) in a notebook. Code should be as minimal as possible to demonstrate the ease of use. Where needed indicate step that will be enhanced in the future, like tiling.
  2. Start the job manager, but from a dataframe that already has finished jobs, let it run, show jobs being created in openEO editor. Show finished jobs in editor.
  3. Show visual progress monitoring in Jupyter or whatever tool is suitable. A map with bounding boxes of jobs + status color would be nice here.
  4. Show data in object storage, maybe using stac browser?

13.3.4 Preparation

The following table outlines the prerequisites for preparing the demo:

Action Responsible
Prepare the script to execute on the CDSE environment (including upscaling tools). VITO