Hosting Solution

[jonaschlegel]

Now that INDIGO is coming to an end, we have to decide on the final hosting solution for the online platform, which will ensure a 2D and 3D graffiti visualisation. We have to decide between a dedicated server solution and a managed platform, and consider our technical, budgetary, and maintenance needs.

Motivations:

• tech requirements
  • storage of 3D model data and orthographic photos
  • fast access for users not only in Vienna, but also around the world
  • Integration with CesiumJS
• budget: around 1000€ (since the project is concluding soon there will be no further budget)
• limited in-house technical experience ongoing, since project is concluding

Options:

A. Hetzner Server:

• Offers autonomy and control over the platform.
• Requires hands-on management, potentially leading to additional costs (job position).
• Integration with our tools would need manual setup and maintenance.
• Dedicated server solution but might exceed our budget in the long run due to unforeseen maintenance costs.
• Current cost calculations: 1000€ for one year, not including the maintenance. B. Vercel/Cesium ion Combination:
• Managed solution that takes care of many technical issues.
• Seamlessly integrates with our tools, especially CesiumJS for 3D content, as they also include streaming and storage of the 3D models.
• Cost-effective (at least the free versions), especially considering our budget constraints.
• Offers specialised 3D content management through Cesium ion, aligning with our tech requirements.

Proposal:

Given our specific needs and constraints, I propose we opt for the Vercel/Cesium ion combination. This solution not only aligns well with our technical requirements but also offers a cost-effective and low-maintenance platform. While it has some limitations, especially on the free tiers, the ease of use, integration capabilities, and 3D content management make it a compelling choice.

Alternatives considered:

We did consider going with the Hetzner server, which offers more autonomy and control. However, the potential challenges in terms of setup, maintenance, and additional costs make it less ideal for our situation. While it provides a robust solution, the technical demands and potential budget overruns make it a riskier choice for Project INDIGO.

[jonaschlegel]

After careful consideration, we decided to go with Vercel and Cesium Ion for a first version.