Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - SESAME (Small cEllS coordinAtion for Multi-tenancy and Edge services)

Teaser

\"Mobile data traffic and services, fueled by new demanding personalised applications, proliferate at an immense rate, radically increasing the demand in infrastructure resources so as to keep user experience at a satisfactory level. Up to now, this ever-increasing demand has...

Summary

\"Mobile data traffic and services, fueled by new demanding personalised applications, proliferate at an immense rate, radically increasing the demand in infrastructure resources so as to keep user experience at a satisfactory level. Up to now, this ever-increasing demand has been fulfilled by the continuously evolving technological framework (3G, 4G), which has offered improved coverage and capacity as well as improved resource usage. However, the long anticipated 5G model needs to involve a paradigm shift, that is to establish a next generation network framework achieving reliable, omnipresent, ultra-low latency, broadband connectivity, capable of providing and managing critical and highly demanding applications and services. The fresh, groundbreaking advances in the field are expected to enforce revolutionary changes in network infrastructure and management, offering the power to align with a demanding set of diverse use cases and scenarios. For all these purposes, the 5G scene needs to couple fast connectivity and optimised spectrum usage with cloud networking and high processing power, optimally combined in a converged environment.
Specifically, one of the envisaged key elements of the 5G technological framework is the capability to deliver intelligence directly to network’s edge, in the form of virtual network appliances, jointly exploiting the emerging paradigms of Network Functions Virtualisation (NFV) and Edge Cloud Computing. 5G network infrastructures need to offer rich virtualisation and multi-tenant capabilities, not only in term of partitioning network capacity among multiple tenants, but also offering dynamic processing capabilities on-demand, optimally deployed close to the user. The potential benefits from such an approach trigger the interest of Communications Service Providers (CSPs) such as Mobile Network Operators (MNO), Mobile Virtual Network Operators (MVNO) and Over-The-Top (OTT) content and service providers, allowing them to \"\"gain\"\" an extra share in the network market by pursuing emerging business models. Following this direction, novel business cases will produce added value from any kind of infrastructure or application that has the potential to be offered \"\"as-a-Service\"\".
While the virtualisation of the communications infrastructure (core/edge segments and access points/macrocells) has been extensively studied by several industry and research initiatives up to now, the applicability of this paradigm to Small Cell (SC) infrastructures has received so far very limited attention. The Small Cell concept has become pivotal in today’s 4G access; Small Cells provide improved cellular coverage, capacity and applications for homes and enterprises as well as dense metropolitan and rural public spaces. Without any doubt, their role is \"\"crucial\"\" for providing services in populated areas like stadiums, shopping malls, concert venues, and, generally, places with (tactic or sporadic) high end-user density. In such cases, normally each telecom operator deploys their own infrastructure, acting complementary to the macro cell network. Normally, Small Cell provisioning requires a number of time and money consuming procedures as e.g., provisioning of installation site, power supply and so on. Operators must also face the costs of establishing dedicated, high-capacity backhaul connections, not to mention radio resource management and interference mitigation techniques, all translating to extra costs and efforts. However, this static approach based on the ownership of the physical Small Cell infrastructure not only increases operators’ CAPEX and significantly hampers business agility, but also is it unable to cope with dynamic scenarios. For example, one should consider the case where sporadic flash crowd events arise not only at predefined venues (e.g., shopping malls, urban areas, stadiums, etc.) but also at arbitrary areas with minor infrastructure in place, resulting in traffic overflow and signal outage. In ord\"

Work performed

The SESAME work plan is tailored in several well-defined WPs, all addressing the main objectives of the Project. The overall work to be conducted as part of SESAME has been organised in eight WPs:
Project Coordination activities have been concentrated into a separate work-package, that is WP1 (Project Management, Risk & Quality Management, (Task 1.1, Task 1.3) and Technical Coordination (Task 1.2)), which will run throughout the entire duration of the Project.
The dissemination and communication, and exploitation activities also constitute a distinct work-package, that is WP8 (Dissemination, Standardisation, Exploitation and Training Activities), which will also be active throughout the Project. It can be assumed that WP1 and WP8 run in parallel and “interact” with the technical activities at every stage of the Project.
The categorization of the work into WPs follows the natural progress of the Project. The first phase is a requirements analysis, use case definition, high-level and detailed specifications, technical characteristics of the system, as well as Proof-of-Concept (PoC) integration inputs preparation. This work is concentrated in WP2 (Requirements, Specifications and Architecture).
Following the definition and specification phase, the implementation of SESAME components architecture is carried out in four work-packages, all running in parallel.
WP3 (Small Cell Design and Implementation) and WP4 (Light DC Design and Implementation) are dedicated in prototyping and implementing the CESC main components.
WP5 (Infrastructure Virtualisation and Management) studies virtualisation strategies and develops the management plane components, including CESC Abstraction Model and VIM implementation. WP6 (Orchestration and Service Level Management) realizes the orchestration components, optimizes the VNFs to be evaluated within SESAME and provides the service management framework.
The outputs of WP3, WP4, WP5 and WP6 are integrated, in-lab tested, and validated within WP7 (Integration, PoC and Evaluation). In WP7, the entire SESAME platform is integrated, deployed and validated as a complete system and also as separate components in the participants’ labs. To that end, the realisation of several use case scenarios is also envisaged as part of WP7.
During the first (1st) Reporting Period (RP), the work performed, in total, was fully aligned to the structuring and/or the detailed context of each dedicated WP.
All respective deliverables and milestones scheduled for the 1st RP have been fully realized, without significant delays or with short delays, per case.
Although the SESAME Project is still in the first year of its intended activities, the proper “coverage” of all scheduled works together with all other necessary actions provides the certainty that all expected objectives and/or impact are to be fulfilled in the forthcoming 2nd RP.



The main (conceptually-oriented results) achieved so far -also included within the submitted SESAME deliverables- are summarised as follows:

Definition and specification of the system architecture and interfaces for the provisioning of multi-operator Small Cell (SC) networks, optimised for the most promising scenarios and use cases.
From the early beginning of SESAME, high-impact use cases have been identified. These will “drive as a whole” the definitions and specifications of SESAME architecture and the high level architecture of the subsystems and of the proposed frameworks. An open architecture and open APIs (Application Programming Interfaces) will be followed in order to make network operators (virtual or not) trust the multi-tenancy structure. The specific following achievements have been performed:
- The proper definition of the system use cases and related requirements.
- The definition and establishment of the overall system architecture as well as of any related interfaces.
- Specification of the CESC components.
- Specification of the Infrastructure Virtualization

Final results

A fundamental component of SESAME is the virtualisation of Small Cell and their utilisation and partitioning into logically isolated “slices”, offered to multiple operators/tenants. The main aspect of this core innovative feature should the capability to accommodate multiple operators under the same infrastructure, satisfying the profile and requirements of each operator separately. This should significantly reduce the cost of deployed infrastructure (i.e., cost of ownership, maintenance, etc.), since hosted Small Cells can be treated as an operating resource instead of a capital expenditure. Under this perspective, the creation of neutral host solutions comes to address also the economic viability of telecom investments.
Also, up to now, network equipment deployed at the edge and access network part had well specified “hard-wired” functionalities that were not possible to be repurposed. With the advent of Cloud Computing, Software Defined Networking (SDN) and Network Function Virtualization (NFV), the idea to have general-purpose computing and storage assets at the edge of mobile networks has matured. In this direction, new industry initiatives have already introduced the concept of Mobile-Edge Computing (M,EC) and the related key market drivers. To enhance further the virtualisation capabilities of the Small Cell deployment and to include not only network capacity resources but also edge processing capabilities, a micro scale virtualised execution infrastructure is proposed by SESAME, in the form of a “Light Data Centre” (Light DC). The Light DC will be designed in order to build a clustered infrastructure with high manageability and will be optimised to reduce power consumption, cabling, space and cost. This aspect not only will optimise end-users’ experience with respect to performance issues, but also, will it give birth to new monetisation chances, i.e., it will provide an ecosystem with novel services residing inside the network infrastructure.
To realise SESAME’s vision, “Cloud-Enabled Small Cells” (CESCs) will be designed, developed and implemented, in order to offer access to network capacity coupled with mobile edge computing resources in a single device. These resources will be offered on-demand to Communications Service Providers (CSPs), profiling both access and edge computation resources to satisfy the specific CSPs’ needs.

Figure 3 illustrates a possible scenario of applicability of the novel framework proposed by SESAME. The proposed infrastructure can be deployed, for example, in specific high traffic demanding areas, such as downtown business regions, dense urban areas, stadiums, shopping malls, etc. Then, CESCs could provide to operators and service providers the required capacity to serve their users’ needs. From the perspective of service provisioning, the proposed approach can be used to provide edge cloud capabilities and enable accelerated services, content and application due to the increased network responsiveness. Operators may provide the network’s edge (i.e., the Light DC) to third party partners, allowing the rapid deployment of cutting-edge services to users and enterprises, translating to added value and creating opportunities for vendors, service providers and operators by enabling them complementary and advantageous positions. Besides, the Light DC will enable the rapid on-demand deployment of cutting-edge network services in the form of Virtual Network Functions (VNFs) – such as data processors, security appliances, proxies, media transcoders, Machine-to-Machine (M2M) gateways etc., close to the mobile nodes. Locating virtual service processing nodes closer to users reduces latency, improves throughput, and reduces traffic in the network core.

The realisation of the SESAME framework has to go through a number of specific needs and requirements. Challenges relate to the possibility to “slice down” a single CESC (or a CESC cluster) in order to “furnish and

Website & more info

More info: http://www.sesame-h2020-5g-ppp.eu/.