The Crenger.com platform architecture includes the following elements. 
Application server
The application server (WildFly) provides a lightweight, high-performance runtime environment for building, deploying, and managing enterprise Java applications. It serves as a fully featured container that handles all the underlying infrastructure needs—such as security, database connectivity, and messaging—so developers can focus strictly on writing business logic. It gives developers access to a standardized suite of enterprise tools (like Jakarta RESTful Web Services and JSON Processing) without locking them into a single vendor's proprietary technology.
Web server
The primary purpose of the web server is to intercept and process all incoming HTTP/HTTPS network traffic before routing it to enterprise Java applications. An application server handles heavy internal business logic, while the integrated web server acts as the frontline gatekeeper.
SQL server
The purpose of the SQL database server (such as PostgreSQL) connected to the application server is to provide a persistent, reliable, and structured transactional data store for enterprise applications.
While the application server manages execution, threads, and web requests, it is inherently stateless. It relies entirely on the external SQL database server to save and maintain information across user sessions and system restarts.
In enterprise systems, multiple database actions must succeed or fail together. The SQL server coordinates these actions. It acts as the transactional muscle, guaranteeing that data remains consistent, isolated, and safely written to disk even if the application server suddenly crashes or loses power.
Graph database server
The primary purpose of a graph database (such as Memgraph) is to unify data into a structured knowledge graph underlying GraphRAG (Graph-Retrieval Augmented Generation) data pipelines of AI systems and agentic workflows.
In Agentic AI—where autonomous AI agents plan, reason, use tools, and make decisions over time— the graph database serves as the real-time memory and state management engine, because standard LLMs are stateless – they don’t store information between prompts. Agent memory is divided into three layers (episodic, semantic, and procedural memories ), all of which map natively to a graph database.
Project IDE
Project Integrated Development Environment (IDE) is a Java enterprise application for desalination project engineering and management. IDE features a loosely coupled, easily extensible architecture that ensures high scalability, maintainability, security, and backward compatibility. Project IDE is primarily for plant designers who understand the project’s inner workings and workflow.
Web project browser
Web Project Browser is a collection of services built with the open-source Vue.js framework. Its popularity is explained by several unique features. It makes UI development more efficient, modular, and maintainable. Next, it simplifies the creation of dynamic user interfaces and Single-Page Applications. Finally, it offers a declarative, component-based programming model that runs on standard HTML, CSS, and JavaScript.
The web project browser is primarily for those who only use the project information (managers, OEMs, subcontractors, and customers).
OT simulator
OT (Operator Training) simulator is a Java enterprise application that links a real-time dynamic process model with an emulated industrial control system. It replicates HMI graphics, alarms, and control consoles identically to the real plant control room environment.
The OT simulator software is auto-generated from the FEED package produced with the PlantDesigner package. Additionally, it serves as a synthetic data source for AI system training.
API
An Application Programming Interface (API) is a software intermediary (hiding the complexity of an underlying system) that allows two distinct applications (like an application server, a mobile app backend, or a web application) to interact, share data, and communicate using a standardized set of rules and definitions. API handles real-time data streaming and standard request-and-response data exchange using a predictable structure (like REST), standardized formats (primarily JSON), and HTTP methods. API implements standard authentication and authorization mechanisms. API is stateless: every single API request from a client must contain all the information and context necessary to understand and process it.
Mobile Apps
Mobile apps offer superior performance, offline functionality, and deep integration with device hardware as compared to web applications. Apps can send direct alerts (aka push notifications) to a user's screen to bring them back, which heavily boosts the worker’s situational awareness. Next, mobile apps connect directly to phone hardware like cameras, GPS, Bluetooth, and biometric scanners, which substantially extends the automation scope. Finally, apps can store data locally on the phone, allowing users to view content or complete tasks without an active internet connection.