SRCNet common services node computing profile for v0.1

The SRCNet will be composed of SRCNet nodes that will have common services running on, called from now on, connecting servers (usually virtualised) and that will be connected to, called from now on, computing servers (usually part of a cloud)

For v0.1, connecting servers will have the following services among others:

• Science Platform Presentation Layer (webserver)
• Common data-related services
• Services Discovery-related services
• Interactive analysis interface
• Orchestrator to deploy, manage and scale services
• Other local data parsing and visualisation services (SODA, visualisation services, etc)

For v0.1, computing servers should have general computing and storage capabilities able to analyse the data and closely handle the needs of the science verification phase for the first SRCNet deployment

An estimation of the profiles of connecting servers and computing servers per SRCNet node that will participate in the v0.1 operational deployment is needed to select the initial distribution (a minimum of 4 SRCNet nodes will be selected)

For connecting and computing servers, this effort will define aspects like:

• Computing power (CPU, GPU)
• Local storage
• Memory
• Architecture
• Bandwidth between nodes
• Bandwidth between disks and computing units
• etc

This feature will define a basic infrastructure architecture of all the servers required for a running SRCNet node for version v0.1

Also, the FTE support required to keep the nodes up and running, provide deployment support, operational activities and maintenance and profiles of the engineers will be also required