appIDALICE / iudx-deployment

Installation and setup scripts for single and multi node (clustered) IUDX services.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

IUDX

iudx-deployment

Installation and setup scripts for single and multi node (clustered) IUDX services.

Components

The IUDX system will consist of the following services and servers:

  • API Server: Implements the Catalogue and Resource Server APIs (Vert.x HTTPs server)
  • Database server: An IUDX Server
  • Data broker Server: An IUDX Server
  • Authentication and Authorization Server: An IUDX Server
  • Database Service: Connects with the IUDX Database (A Vert.x Service)
  • Databroker Service: Connects with the IUDX Data broker (A Vert.x Service)
  • Subscription and Callback Service: Connects with the IUDX Database and Databroker (A Vert.x Service)
  • Validation Service: Validates the catalogue item (A Vert.x Service)
  • File Service: Connects with the IUDX Database, Databroker and File server (A Vert.x Service)
  • Authentication and Authorization Service: Connects with the IUDX Authentication Server (Vert.x Service)

Deployment Architecture

Note: For simplicity, all the modules are not mentioned in the figure.

As a design choice on the architecture, keeping scalability for microservices in mind we chose the Service Mesh Architecture for Catalogue Server and Resource Server. In a service mesh architecture, each microservice is a well-defined module that can be containerized and discovered using service discovery. The orchestration of the services can be such that data-intensive modules are residing closer to the database which also helps in better response times, limits the bandwidth and reduces the cost. Also, it helps in scaling of a specific microservice at ease.

Features

  • The system uses an overlay network and supports manual scaling using docker run.
  • Hazlecase with Zookeeper is used as our cluster manager.
  • Monitoring of the APIs, Services, Containers, Nodes are done using Micrometer, Promtail and Node exporter.
  • Metrics are pushed to Loki, Prometheus and visualized using Grafana.

Contributing

We follow Git Merge based workflow

  1. Fork this repo
  2. Create a new feature branch in your fork. Multiple features must have a hyphen separated name, or refer to a milestone name as mentioned in Github -> Projects
  3. Commit to your fork and raise a Pull Request with upstream.
    A detailed instructions present here.

About

Installation and setup scripts for single and multi node (clustered) IUDX services.

License:MIT License


Languages

Language:Mustache 53.2%Language:Shell 34.7%Language:FreeMarker 5.5%Language:Smarty 2.7%Language:Python 2.2%Language:Dockerfile 1.0%Language:CSS 0.7%Language:Jinja 0.1%