Service Requirements Gathering

SIG Greenhouse

Nicole Harris

Developer community waning within the NREN world.  Lots of reliance on OSS but no strategy for long-term support for the software, very few NRENS have development strategy in place alongside procurement strategy, difficulties with maintaining developers.

Work on Docker recipes and approaches. 

SIG Multimedia (was WebRTC)

Peter Szegedi

Open Educational Content: make them accessible and re-usable. 

In-context based real-time communications. Application development based on the  WebRTC protocol. Rel time traffic (audio, video, data).

Make the GÉANT TURN Service pilot global (nodes in Americas, Asia Pacific, Africa).

SIG SCOPE

Nadia Sluer

SIG-ISM / WISE

Sigita Jurkynaite, Alf Moens (chair)

• What new service / development requirements do you see for NRENs in the next 3 - 5 years?
• Information security measurement (ISO27004:2016)
•  What are the priorities and specific ideas for innovation in these areas?
• How can such measurement help the NRENS to establish a more complete and integrated ISMS
•  What services and capabilities not currently offered by GÉANT will be needed i n 3-5 years?
• Information Security measurment
• 5G development v.s. WiFi and Eduroam

SIG-Marcomms and SIG MSP

Laura Durnford, Gitte Kudsk (DeIC), Maria Ristkok (EENet), Lonneke Walk (SURFnet - chair)

Laura Durnford / Magda Haver, Walter van Dijk (SURFnet), Mark Tysom (Jisc), Harri Kuusisto (CSC-Funet), Alberto Perez (REDiris), Martin Bech (DeIC - chair)

Presentation by SURFnet:

• Envision needing more services focused on campus domain, ICT services etc and so GEANT would need to help improve national service delivery and economies of scale.
• Current GEANT portfolio reflects tech- driven services. IN NL have started to facilitate coalitions of the wiling – Ict managers  etc to help develop and launch new services, result in extension of the portfolio. So want demand aggregation, joint tendering, tender management as something GEANT can do to support.
• Need staff available to facilitate large-sclae procurement, operation management in trust, security & cloud – GEANT leans on NRENs for running T&I and security services, would be good if its staff could do that. Also if could support the ICT mangmnt of orgs such as NATO, ESA, UN that need GEANT and NREN resources.
• GEANT academy to help NRENs acquire skillsets.
• Automation of network services.
• Just entering campus domain with service that allows us to manage their wifi – would be good to use a control plane to include campus domains.
• Clouds – many campuses want to get rid of ICT depts. And use commercial solutions via NREN as broker. See commercial services where we know our customers want to be able to get them from us.
• Want white label applications or software so can incorporate into NREN own portfolio
• Open protocols and software are a must
• Standardisation where possible.

Presentation by DeIC

• Want GEANT to be the vehicle for everything we want to do in common among the NERns
• Want cost reductions and opp for NRENs to make money supplying stuff to the rest of the community
• Abolish the GENAT backbone and have it base on NREN connections and cross-border fibres
• Network-side DDoD mitigation service
• Joint procurements
• NREN-to-NREN service provisioning
• Enable eduGAIN to support ad hoc federations.
• Orchestration of point-to-point connections through open APIs
• Users, funders etc should come up with KPIs and we come up with the numbers.
• Services should be white-labeled in several levels.
• As a community should have a complete portfolio - if a new country emerged should have an ‘NREN-as-a-service’ offering.
• funding model for NERNS has big impact. RedIRIS centrally funded, DeIC more in competition with others.

SIG-NGN

Brook Schofield, Rob Evans (chair)

Question posed to the entire mailing list. Awaiting comments.

Feedback in the STF meeting collected by Annabel/Nathalie might be relevant.

SIG-NOC

Charlie van Genuchten, Brian Nisbet (chair)

• There is a trend towards 24/7 support.
  
  • With the increase of international cooperation and exchange, clients have users from different time-zones and different countries, so they expect their students and employees to be able to have access to the network from any place at any time.
• There will be a need for more managed services :
  • Radius
  • Authentication services
  • Federated sign on
  • Things that have been installed at one point in time and if they break, there is no one to fix them, especially in this age of automation.
• More and more networks are being merged, which will bring new challenges in the future.

Input CSUC

• What new service / development requirements do you see for NRENs in the next 3 - 5 years?

You already offer most of the services NREN might require now. One idea would be offering Assessment to NREN and its connected entities (I think Jisc already does something). As there are so many different services, so many clouds, so complicated laws...and the network is seen as a commodity, but is more complex than that. Big universities or entities usually have good teams and technicians, but small institutions can't afford to have many people and they need assessment. This assesment could be in many fields, like:

- Deciding what services should go to the cloud and what services don't need a cloud transformation
- General Data Protection Regulation
- Network configuration
- Network monitoring (like Network Monitoring as a Service)
- Teaching support innovation
- Virtual Campus creation, development and deployment
- Big data

Maybe Géant could offer this assessment, maybe each NREN could specialize in one or several items.

• What are the priorities and specific ideas for innovation in these areas?  What services and capabilities not currently offered by GÉANT will be needed i n 3-5 years?

Innovation for e-learning, teaching services to foster online teaching, master classes, e-portfolios.

What is the potential impact of new technologies such as orchestration, automation and controlling applications?

Applications control everything, but automation and orchestration are essential to keep these applications running and to have all our network and systems equipment (logical and physical) items under control. Monitoring is always healthy, as a key for fast resolution of incidents and to anticipate future issues. Not only for the network.

• How well do we do at NREN commitments early in the process of service device development?  The use of services, facilities and even the network varies a lot from NREN to NREN:  Should new services or new service features be developed only after a certain minimal number of NRENs express their commitment to offer/use them for their business?

And this is exactly what reminds me on what we do. We are a consortium of 10 universities. We ask them what they would like to get from us, but before offering a service or even beginning with all the process to offer it, we need at least 3 of them committed to contract or use the service. This means 30%. It works well for us. It could work well for you.

• Do you think we should make an effort to serve communities outside of the R&E community - following the ECs evolving priorities and flagship projects?

I would involve enterprises if they are related to research and education. We do it here, although these enterprises do nos get their internet connection trough RedIRIS or Géant, but through a commercial provider. They have an excellent connection to the R&E institutions in Catalonia, because they have projects or collaborate with them.
On a Spanish level, RedIRIS uses the network for the emergencies network. It could be a good use of it.

• Do you see any significant future changes in the NREN business models affecting what they want to get from GÉANT Ass/Ltd. E.g Importance of SLAs and KPIs, use of non-commercial software from GÉANT Ass/Ltd, other projects or research groups

As a philosophical idea, NREN (and GÉANT) should be flexible enough to adapt to the needs of the connected institutions and to the changes on the technology and the network. Sometimes, they may be too tied up because they belong to public sector institutions, and public sector things are usually quite inflexible. NRENs must have the agility to accommodate to changes and be able to offer a good service despite crisis or problems. It has to motivate its customers and users to use the network for research, like a catalyzer.

SIG-PMV

Kurt Baumann

SIG-TNE

Esther Wilkinson, Sigita Jurkynaite

TF-CSIRT

Nicole Harris, Sigita Jurkynaite, Baiba Kasina (chair)

TF-RED

Sylvia Kuipers + (TBC)

TF-Storage / SIG-CISS

Peter Szegedi, SC members

SIG-CISS work item #2 ;) -- transnational compute-job mobility in the form of docker containers that can be moved at will between different NREN cloud stacks, supervised probably by some sort of meta-orchestrator, and ideally with the collection of already-prepackaged workflow containers available in an "NREN science app store".

Federation of data storage with possibilities to run computations close to the data.

• licensed/unlicensed spectrum for education and research purposes.
• the E-SIM for uniform access to unwired services and other cool related stuff.
• Internet of Things: LoRaWAN infrastructure (or others in the licensed/unlicensed spectrum band, e.g. 802.11ah/NB-IoT).
• LTE-U.
• Internet of Things: "Smart Campus".
• Indoor Location Based Services.
• Fixed/Mobile integration on campuses, helping institutes purchasing the right "mobile service".
• Supporting/pushing Wifi-calling at the campus.
• DAS-networks: still needed, or obsolete?
• Robust and redundant unwired network deployments (at campus) for emergency situations.
• 5G projects and developments.
• Wireless service (and beyond?!) by NRENs: do's and don'ts.
• Cooperating with initiative such as WIFI4EU (or FON, iPASS, Boingo...).
• developments and next steps for govroam internationally.

• Internet of Things. 
• Data (with EUDAT? or RDA?)
• Education / Humanities (with EUNIS?).
• Campus Middleware (Internet2 / TIER).
• Extending TCS (ACME Protocol Support + Lets Encrypt)
• Usability.
• 5G stuff.
• Legal.