SKOPE Deployments

SKOPE is a basic microservice architecture with with each service running within container. At the center of the backend services is Clowder, a Data Management System (DMS) designed to manage large collections of scientific data. Clowder can be extended using addition microservices known as extractors. Like many microservice system, Clowder depends on RabbitMQ to handle notifications of new data sets, file uploads, or metadata changes.

Deploying SKOPE

The SKOPE is deployed as a stack of services from a docker-compose.yml file. Deployment with docker-compose.yml allows for easy transition from development to staging to production.

The basic system changes required to deploy SKOPE locally are minimal and shown below. Installation and deployment using existing SKOPE data comes down to the following 3 steps:

1. prep the local system
2. copy data from the existing SKOPE system
3. start the SKOPE services

For detailed instructions for preparing a new server to run SKOPE and getting existing data, see the instructions in the SKOPE server config.

Starting SKOPE

This section is from the skope-deployment repository but repeated here for convenience. See the skope-deployment instructions for the most recent changes.

1. clone the GitHub repository
2. change to staging branch
3. use docker-compose to run start services

$ git clone https://github.com/openskope/skope-deployment.git
...
$ cd skope-deployment
$ git checkout staging
...
$ sudo docker-compose up -d
...