Welcome to CRISalid Deployment Documentation

❔ What is CRISalid?

CRISalid is a modular set of open-source components designed to support the development of an institutional Current Research Information System (CRIS).

It is built to be message-oriented, modular, and loosely coupled. Each component communicates with others through asynchronous messaging (via RabbitMQ), while also exposing classical REST or GraphQL APIs to allow service-oriented approaches.

💡 Core Principles

Free and Open-Source: All components are licensed under GNU GPL-compatible licenses and freely available from GitHub.
Cloud-native: All components are continuously built and published as Docker images, and the whole system is designed to be deployed in cloud environments. You can find them on our DockerHub organization.
Modular Architecture: Not every institution needs all components. You can deploy only the ones you need and integrate them with your existing systems for the management of research-related information.
Standards Compatibility: While CRISalid components are not natively semantic, the data model is “ontology driven” and designed to be compatible with the new CERIF 2 standard. This ensures long-term interoperability and facilitates research information openness, in line with initiatives such as the DORA declaration.
Institutional Integration and Best Practices: Our approach follows the recommendations outlined in the OCLC Research Report (2021) on Research Information Management in the United States, (see the “Recommendations for Institutional Leaders” section). We avoid creating data silos: CRISalid is not a standalone tool but a subsystem within the broader institutional information system. We emphasize interoperability, rely on Persistent Identifiers (PIDs), and promote integration with existent systems like people and structure directories and project or publication repositories at local, national and international levels.

🧩 Main Components

CRISalid Bus
A shared RabbitMQ message broker used for real-time communication between components.
Graph Database
A Neo4j database that hosts the institutional Knowledge Graph, which acts as the single source of truth for the entire system.

Access to Neo4j is not direct: it is mediated through:
- crisalid-ikg: A FastAPI middleware that maintains and synchronizes the graph.
- crisalid-apollo: A GraphQL API gateway exposing the graph to frontend and analytical services.

🛠️ Deployment Options

This documentation provides two main deployment strategies:

🧪 Development & Test Deployment

Use this if you’re:

Working on a specific CRISalid module
Testing integration locally
Evaluating CRISalid for your institution

We provide:

Docker Compose examples
Manual installation instructions (e.g., for deploying on virtual machines)

→ Explore Development & Test Deployment

🚀 Production Deployment

For institutional or production-grade deployments.
While Kubernetes is the primary target, it is not required — you may still use Docker or manual setups if needed.

We provide:

Shell scripts to generate Kubernetes manifests
Manual installation instructions for production environments

→ Explore Production Deployment

📚 Documentation Structure

The documentation is organized as follows:

System Overview: Understand CRISalid’s components and how they interact.
Development & Test Deployment: Step-by-step guides to run CRISalid locally.
Production Deployment: Resources for full-scale deployments.
Technical Requirements: Infrastructure and dependencies checklist.
Security Considerations: Best practices for securing your CRISalid deployment.

Use the navigation menu on top to browse.