While shared services, like the data and application catalogs are owned and operated by the MDPS team for all users of the MDPS system, venue services are single tenant services used only by the project team. MDPS venue services are easily deployed, upgraded, and destroyed from the venue Management Console.
There exists an Multi-mission operations team to configure and deploy resources into the project venue. This team is responsible for ensuring that a given capability is available and performing, and trouble shooting any issues with underlying infrastructure, connectivity, and permissions. The MMO team is not responsible for algorithm authoring, operational science product generation, or optimization of algorithms.
The following venue services are available for use in a project venue. Most of these resources have many options, such as underlying instance types, scalability, and naming conventions.
- Processing Service - a cluster capable of executing application packages at scale.
- JupyterHub - an environment for algorithm development through an interactive development environment
- Data Bucket - the ability to create a new, persistent bucket for cloud data storage. Future functionality will include backup, versioning, and other project specific options.
- Science Data Analysis Platform (SDAP) environment (coming soon) - a platform for analyzing large scale data collections.
The management console is how services are deployedm, monitored, and managed within a given venue. Projects are allowed to use this interface, or can work with the MMO to schedule deployment, maintenance, and upgrade requests.