Periodic Reporting for period 3 - 5GPagoda (A network slice for every service)
Okres sprawozdawczy: 2018-07-01 do 2019-06-30
The top objectives of 5G!Pagoda are the development of a scalable 5G slicing architecture towards supporting specialized network slices composed on multi-vendor network functions, through the development of a scalable network slice management and orchestration framework for distributed, edge dominated network infrastructures, and convergent software functionality for lightweight control plane and data plane programmability and their integration, customization, composition and run-time management towards different markets in Europe and Japan. The project also aims to align the views of Japanese and Europeans about the 5G mobile system.
Towards this end, in the first year of the project, the 5G!Pagoda consortium has defined and analyzed an architecture that consists of multiple components that meet these objectives, taking into account slicing over multiple technological and administrative domains. The consortium has also enabled the concept of end-to-end mobile network slicing, an important step towards the autonomic orchestration of resources to create end-to-end network slices across multiple domains and in manner customized to different mobile services.
In the second year of the project, the consortium has focused its efforts on devising new designs for lightweight control plane functions, new schemes for deep data programmability, new algorithms for the composition of slices and the placement of VNFs. The consortium has also looked into the scalability issues relevant to slicing and proposed new solutions, leveraging the resource pool concept and incorporating the in-slice management approach.
The third year of the project (as well as the second half of the second year) was mainly characterized by the efforts of the consortium partners in enabling different test beds that lead to the creation of different slices for different service types (e.g. IoT service, video conferencing services, and on-demand video streaming services). This goes very well inline with the slogan of the proposal, namely a slice for every service. These testbeds are described in details in WP5 deliverables. Their supporting algorithms and mechanisms are detailed in WP3 and WP4 deliverables.
Several dissemination and standardization activities have been also carried out to promote the finding of 5G!Pagoda. The research work conducted by the 5G!Pagoda is followed by a growing community of researchers and industrials in Europe, Japan and many other parts of the world. Some of the 5G!Pagdoa approaches with regard to multi-domain slicing have been adopted by other research consortia and an important impact on the scientific community is expected.
In the second year of the project, the consortium has focused its efforts on devising new designs for lightweight control plane functions, new schemes for deep data programmability, new algorithms for the composition of slices and the placement of VNFs. The consortium has also looked into the scalability issues relevant to slicing and proposed new solutions, leveraging the resource pool concept and incorporating the in-slice management approach.
The third year of the project (as well as the second half of the second year) was mainly characterized by the efforts of the consortium partners in enabling different test beds that lead to the creation of different slices for different service types (e.g. IoT service, video conferencing services, and on-demand video streaming services). This goes very well inline with the slogan of the proposal, namely a slice for every service. These testbeds are described in details in WP5 deliverables. Their supporting algorithms and mechanisms are detailed in WP3 and WP4 deliverables.
Several dissemination and standardization activities have been also carried out to promote the finding of 5G!Pagoda. The research work conducted by the 5G!Pagoda is followed by a growing community of researchers and industrials in Europe, Japan and many other parts of the world. Some of the 5G!Pagdoa approaches with regard to multi-domain slicing have been adopted by other research consortia and an important impact on the scientific community is expected.
In comparison to the state of the art, the consortium was the first to propose and analyze an architecture for slicing across multiple administrative and technological domains. The work conducted here has served as a reference for many other research consortia. In addition, the consortium proposed new design patterns for lightweight control plane functions, new schemes for deep data programmability, new algorithms for the composition of slices and the placement of VNFs, and new solutions to cope with scalability issues in network slicing, leveraging the concept of resource pool and also in-slice management.
The consortium developed a number of orchestrators for the creation of different types of slices, demonstrating the following proofs of concept:
1. Demo for end-to-end orchestration of mobile network slices including RAN slices; featuring RAN slicing and multi-technological domains;
2. Demo for IOT end-to-end network slices orchestration; featuring multi-technological domain orchestration, and IOT-oriented orchestration; and
3. Demo for ICN/CDN end-to-end network slices orchestration; featuring multi-domain orchestration and stitching process.
By the end of the project, the partners have developed the sx following exploitable assets:
• Implementation of the 5G core network to support small messages and data plane diversity led by FOKUS
• Network Service Planner led by Aalto
• Programmable RAN open source OpenAirInterface led by EURECOM and Orange
• Deep data plane programmability system (granular fine grain traffic control) led by NESIC and UT
• Resource Pool for Scalable Orchestration led by HITACHI
• CDN/ICN Slicing as a Service led by Aalto and Waseda
The findings of the project have been so far disseminated through different channels and shared with a wide community of researchers and industrials in the form of scientific publications, demos, keynotes, tutorials, and panels organized at different venues. Contributions to Standards Development Organizations have been also made.
The consortium developed a number of orchestrators for the creation of different types of slices, demonstrating the following proofs of concept:
1. Demo for end-to-end orchestration of mobile network slices including RAN slices; featuring RAN slicing and multi-technological domains;
2. Demo for IOT end-to-end network slices orchestration; featuring multi-technological domain orchestration, and IOT-oriented orchestration; and
3. Demo for ICN/CDN end-to-end network slices orchestration; featuring multi-domain orchestration and stitching process.
By the end of the project, the partners have developed the sx following exploitable assets:
• Implementation of the 5G core network to support small messages and data plane diversity led by FOKUS
• Network Service Planner led by Aalto
• Programmable RAN open source OpenAirInterface led by EURECOM and Orange
• Deep data plane programmability system (granular fine grain traffic control) led by NESIC and UT
• Resource Pool for Scalable Orchestration led by HITACHI
• CDN/ICN Slicing as a Service led by Aalto and Waseda
The findings of the project have been so far disseminated through different channels and shared with a wide community of researchers and industrials in the form of scientific publications, demos, keynotes, tutorials, and panels organized at different venues. Contributions to Standards Development Organizations have been also made.