×
Archytas
Our platform is designed to offer scalable geospatial processing over an indexed data catalog, enabling users to run computations efficiently while maintaining full control over data access and licensing.
Key Features
- Vendor-Agnostic Geospatial Processing
- Users can leverage specialized processing algorithms encapsulated in abstract, non-vendor-locked libraries.
- These libraries can be executed both on a local machine and in a distributed computing environment, ensuring flexibility in deployment.
- STAC-Compliant Data Catalog
- The platform provides an open-access STAC-compliant data catalog, allowing users to search, discover, and index any kind of geospatial data.
- The only requirement for ingestion is that each dataset must include spatial and temporal metadata.
- Controlled Access to Assets
- While the catalog is freely accessible (after user registration via our company’s Identity Provider (IDP)), access to actual assets is controlled.
- A dedicated license-matching component ensures that users can only access assets if their licenses align with the asset’s defined usage rights.
- Asset authors are responsible for defining and assigning licenses to their data.
- On-Demand Scalable Computing with Dask
- The platform integrates a Dask cluster for scalable, parallel computation.
- This cluster is available upon request and must be booked via email at support.archytas@planetek.it.
Architecture
The platform is structured in a way that resembles the Linux kernel, with two primary layers:
- Core Layer (Infrastructure & Security)
- STAC-Compliant Catalog: Stores and organizes geospatial datasets.
- Scalable Processing Engine: Executes distributed computations.
- License & Permission Management: Controls user access to data assets.
- User Layer (Computation & Execution)
- Python Libraries: Abstract geospatial processing algorithms, deployable locally or in the cloud.
- Algorithm Orchestrator (Prefect): Enables users to define and execute geospatial workflows seamlessly.
