Research database and storage strategy

[BytecodeBrewer]

Goal

Research what ARGUS should store and which database/storage approach fits the project.

Why

ARGUS is moving from live requests and in-memory analytics toward real data workflows. Before adding a database, the project needs a clear decision about what should be stored, why it should be stored and whether local, server-based or cloud storage is the right next step.

This ticket is also intended to build practical SQL and database management skills.

Research questions

• What data should ARGUS store?
  • historical exchange rates
  • fetched raw API responses
  • cleaned time-series data
  • calculated metrics
  • user-selected watchlists
  • generated reports
  • data-source metadata
• What should not be stored yet?
• Should ARGUS start with local storage, server database or cloud storage?
• Which database fits this phase best?
  • SQLite
  • PostgreSQL
  • DuckDB
  • Parquet files
• How should tables/entities be structured?
• How does storage affect future analytics, dashboards and cloud deployment?

Scope

• Compare local, server and cloud storage options
• Compare SQLite, PostgreSQL, DuckDB and Parquet for ARGUS use cases
• Propose a first minimal data model
• Document trade-offs
• Recommend first implementation step

Acceptance criteria

• Storage use cases are documented
• Local, server and cloud options are compared
• At least three storage technologies are compared
• A first minimal data model is proposed
• Recommended first database/storage approach is selected
• Research is documented in docs/
• No database implementation is required in this ticket

Note

Priority: Must