The era of the "Everything-in-PHP" monolith is fading.

As we move through 2026, enterprise applications are increasingly "polyglot." While Drupal remains the gold standard for content management and complex governance, other languages often provide better performance or library support for specific tasks like AI processing (Python) or real-time streaming (Node.js).

Upsun makes this transition seamless by allowing you to define and run multiple, independent applications within a single project cluster using Infrastructure-as-Code.

1. The Multi-App Architecture

On Upsun, a project isn't limited to one application. You can define a fleet of services in your .upsun/config.yaml file. Each "app" lives in its own directory within your Git repository but shares the same environment lifecycle.

Common 2026 Polyglot Stack:

• App A (Drupal 11): Handles the CMS, user authentication, and administrative UI.
• App B (Python 3.12): Runs a FastAPI service for RAG (Retrieval-Augmented Generation) and AI-driven content tagging.
• App C (Node.js 22): Manages a real-time dashboard or a high-concurrency Next.js frontend.

2. Private Networking: Security Without Passwords

The biggest advantage of the Upsun cluster is Network Isolation. * Service-to-Service Communication: Your Python microservice can talk to your Drupal backend via a private internal URL (e.g., http://python.internal). This traffic never touches the public internet.

• No Password Risk: Upsun uses private networking and SSH keys for internal authentication. Because the network is isolated, you don't need to manage or rotate database passwords between your microservices.

3. Independent Scaling and Resource Allocation

Microservices often have wildly different resource profiles. Upsun allows you to "Right-Size" each container individually.

• The Python Worker: Give it 8GB of RAM and 4 vCPUs for heavy data crunching.
• The Drupal Backend: Give it 4GB of RAM and 2 vCPUs for standard content serving.
• The Node.js Frontend: Scale it horizontally to 3 instances during a traffic spike without affecting the backend services.

4. Shared Services: One Database, Multiple Consumers

While the applications are separate, they can share the same Managed Services.

• Shared MariaDB/PostgreSQL: Both your Drupal app and your Node.js app can connect to the same database service, ensuring data consistency across your stack.
• Shared Redis: Use a single Redis instance for cross-application caching or as a message broker between your PHP and Python services.

5. Implementation: The .upsun/config.yaml Setup

Building a multi-app cluster is as simple as defining them in your Git-tracked config:

YAML

# .upsun/config.yaml

applications:

  drupal:

    source: { root: "web" }

    type: "php:8.3"

    relationships:

      database: "db:mysql"

  ai_engine:

    source: { root: "ai-worker" }

    type: "python:3.12"

    relationships:

      database: "db:mysql"

services:

  db:

    type: "mariadb:10.11"

Conclusion: The Power of Choice

In 2026, the most resilient architectures are those that remain flexible. By integrating Python or Node.js directly into your Drupal environment on Upsun, you remove the "Technology Debt" of forcing a single language to do everything. You get the governance of Drupal, the intelligence of Python, and the speed of Node.js—all managed by one unified WebOps workflow.

Is your Drupal site ready to go polyglot?

We specialize in architecting multi-language microservices and AI integrations on the Upsun platform.