Achtung:

Sie haben Javascript deaktiviert!
Sie haben versucht eine Funktion zu nutzen, die nur mit Javascript möglich ist. Um sämtliche Funktionalitäten unserer Internetseite zu nutzen, aktivieren Sie bitte Javascript in Ihrem Browser.

Foto: Judith Kraft Show image information

Foto: Judith Kraft

PICTURE - Programmable Infrastructure Converging Disaggregated Network and compute Resources

Motivation:

The explosive growth of mobile internet traffic attributed to the rapidly increasing number of smart devices forces network providers to concurrently support a large variety of services that can be either stand alone or interact. This introduces the need to transform traditional closed, static and inelastic network infrastructures into open, scalable and elastic ecosystems that can support a large variety of dynamically varying applications and services. This transformation aims to bring new service capabilities to network operators in terms of: i) connectivity for a growing number of very diverse devices, ii) high mobility in heterogeneous environments and, iii) mission critical services currently handled by specific purpose networks, supporting highly variable performance attributes in a cost and energy-efficient manner.

These demanding and diverse requirements cannot be supported by current best effort internet connections or a smooth migration from existing network architectures and technology deployments to next generation solutions. Instead, it requires a paradigm shift to support the upcoming service requests following new service models such as cloud-based solutions, imposing enormous capacity requirements that exceed 1000 times current bandwidth needs, available to 10s of billions of end-devices. In this context, a future proof infrastructure needs to address a wide range of vertical industry services adopting a flexible architecture offering converged services across heterogeneous technology domains deploying unified software control.

To this end, 5G-PICTURE provides a platform that transforms current closed, purposely developed and dedicated infrastructures into open environments where resources can be accessed and shared on demand supporting very diverse services that can vary dynamically in time. The level of service variability will go beyond just variability in the volume of demands and will cover also other characteristics relating to mobility, density etc. This will make the 5G-PICTURE infrastructure suitable to support in essence any service associated with mobile telecommunications networks as well as any other vertical industry through a common infrastructure.

Objectives:

5G- PICTURE follows the generic 5G design principles defined by the 5G-PPP community and designs and implements a 5G platform offering:

  • Support of 5G ICT and vertical operational and end-user services.
  • Integration of heterogeneous network technologies (wireless and optical).
  • Integration of communication and computation technologies.
  • Softwarisation of network and vertical-specific components enabling a paradigm shift from network entities to network functions.
  • Slicing for multi-tenancy and efficient resource sharing.

Work Packages

The Computer Network Research Group of Paderborn University is mainly involved in the following work packages:

Work package 2: 5G and Verticals Services, Requirements and Architecture

This work package identifies the 5G ICT and “Vertical” services (i.e. relevant to other industries outside the telecom ecosystem such as Rail, media etc. supported by 5G solutions) and their related requirements to be considered in the project. It is also responsible for the analysis and benchmarking of the proposed architecture, by investigating its performance in terms of a set of KPIs) as well as its scalability and sustainability potential.

Work package 5: 5G Operating System

To manage and operate complex heterogeneous infrastructures, WP5 integrates SDN and NFV and optionally existing slicing systems by using an instrumented execution container. This WP uses infrastructure resources flexibly, dynamically and efficiently, simplify infrastructure management and deployment of chained services.

Work package 6: Demonstration of 5G ICT and Vertical Use Cases

The aim of this Work Package is to validate the feasibility of the functionalities and the architectural solutions developed in the project and demonstrate them in two vertical scenarios namely the provision of end-user and operational services at smart railway and crowded stadium use cases.

Work package 7: Standardisation, Dissemination and Exploitation

This work package deals with dissemination/communication, standardisation and commercial exploitation related activities.

Publications


Open list in Research Information System

5G OS: Control and Orchestration of Services on Multi-Domain Heterogeneous 5G Infrastructures

S. Dräxler, H. Karl, H. Razzaghi Kouchaksaraei, A. Machwe, C. Dent-Young, K. Katsalis, K. Samdanis, in: 27th European Conference on Networks and Communications (EUCNC 2018), 2018


JASPER: Joint Optimization of Scaling, Placement, and Routing of Virtual Network Services

S. Dräxler, H. Karl, Z.A. Mann, IEEE Transactions on Network and Service Management (2018)

DOI Download
Abstract

To adapt to continuously changing workloads in networks, components of the running network services may need to be replicated (scaling the network service) and allocated to physical resources (placement) dynamically, also necessitating dynamic re-routing of flows between service components. In this paper, we propose JASPER, a fully automated approach to jointly optimizing scaling, placement, and routing for complex network services, consisting of multiple (virtualized) components. JASPER handles multiple network services that share the same substrate network; services can be dynamically added or removed and dynamic workload changes are handled. Our approach lets service designers specify their services on a high level of abstraction using service templates. JASPER automatically makes scaling, placement and routing decisions, enabling quick reaction to changes. We formalize the problem, analyze its complexity, and develop two algorithms to solve it. Extensive empirical results show the applicability and effectiveness of the proposed approach.


Towards Predicting Resource Demands and Performance of Distributed Cloud Services

S. Dräxler, M. Peuster, M. Illian, H. Karl, 2018

Download
Abstract

Understanding the behavior of distributed cloud service components in different load situations is important for efficient and automatic management and orchestration of these services. For this purpose and for practical research in distributed cloud computing in general, there is need for benchmarks and experimental data. In this paper, we describe our experiments for characterizing the relationship between resource demands of application components and the expected performance of applica- tions. We present initial results for predicting the interdependence between resource demands and performance characteristics using support vector regression and polynomial regression models. The data gathered from our experiments is publicly available.


Programmable and Flexible Management and Orchestration of Virtualized Network Functions

H. Razzaghi Kouchaksaraei, S. Dräxler, M. Peuster, H. Karl, IEEE, 2018

Abstract

Supporting the vast variety of network services’ management and orchestration requirements is one of the main challenges that Network Function Virtualization (NFV) is dealing with. While general management requirements such as Virtual Network Function (VNF) resource requirements can be specified by the service developers using service descriptors, specific management operations like VNF-specific configuration cannot be performed by these descriptors. On the other hand, it is inefficient and also very challenging for Management and Orchestration (MANO) frameworks to provide all specific-management operations for every individual network service and their constituent VNFs. To mitigate this issue, we propose the use of service-specific programs called Specific Managers (SMs) that can customize management and orchestration of network services and also extend the capability of MANO frameworks to support per-service management and orchestration. The results of our evaluation show that the higher flexibility and programmability enabled by SMs improve the performance of the service performance and also utilises the service provider resources more efficiently.


Pishahang: Joint Orchestration of Network Function Chains and Distributed Cloud Applications

H. Razzaghi Kouchaksaraei, T. Dierich, H. Karl, 2018

Abstract

Developing cloud applications using a microservice architecture allows their functional blocks to be distributed and deployed on multiple Cloud infrastructures. This enables service providers to mix and match Cloud-based microservices and Virtual Network Functions (VNFs) that are provided by Network Function Virtualization (NFV). Provisioning complex services containing VNFs and Cloud-based microservices across NFV and cloud infrastructures can enhance service quality, reduce latency, and optimise cost. This can be provided by an orchestration system that can handle cross-ecosystem dependencies. To this end, we implemented Pishahang that is a framework for jointly managing and orchestrating virtual network functions and Cloud-based microservices. During the demo, we deploy several complex services to demonstrate features provided by Pishahang to support management and orchestration of complex services.


Generating Resource and Performance Models for Service Function Chains: The Video Streaming Case

S. Dräxler, M. Peuster, M. Illian, H. Karl, in: 4th IEEE International Conference on Network Softwarization (NetSoft 2018), 2018

Abstract

Understanding the behavior of the components of service function chains (SFCs) in different load situations is important for efficient and automatic management and orches- tration of services. For this purpose and for practical research in network function virtualization in general, there is a great need for benchmarks and experimental data. In this paper, we describe our experiments for characterizing the relationship between resource demands of virtual network functions (VNFs) and the expected performance of the SFC, considering the individual performance of the VNFs as well as the interdependencies among VNFs within the SFC. We have designed our experiments focusing on video streaming, an important application in this context. We present examples of models for predicting the interdependence between resource demands and performance characteristics of SFCs using support vector regression and polynomial regression models. We also show practical evidence from our experiments that VNFs need to be benchmarked in their final chain setup, rather than individually, to capture important interdependencies that affect their performance. The data gathered from our experiments is publicly available.


Open list in Research Information System

Further information:

PICTURE

Information about the project:     
Project members:Holger Karl
Hadi Razzaghi
Sevil Dräxler
Project website:www.5g-picture-project.eu
Type:5G-PPP
Started:June 2017
Finished:Active
Contact:Holger Karl

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 762057.

The University for the Information Society