Film Strip (optimised).png

Video Archive

Internal & external presentations from the University of Bristol, University

of Cambridge, Lancaster University & University of Surrey.

Arrow Graphic.png

An ambitious programme geared to create a radically new architecture for the UK’s internet and telecommunications infrastructure

BT Thought Leadership 2022

A new series of linked talks in 2022 by principal academics on the NG-CDI project, hosted by the BT Thought Leadership programme...

Dynamic Risk & Governance
Philip Stiles, Corporate Governance and Co-Director of the Centre for International Human Resource management, Cambridge Judge Business School. 14/06/2022

During the NG-CDI research we have developed a model of risk and governance that links enterprise risk and model risk management. In this talk, Philip takes these ideas forward and looks specifically at the distinction between enduring and dynamic risks. This distinction is being used within BT as part of a project to reframe risk and governance away from a predominantly compliance focus towards a more behavioural, culturally embedded, perspective. In our research we have seen similar transitions in other organisations. We present the implications of this approach for risk and governance by drawing on examples from our use cases and from discussions conducted within BT, as well as further presenting an overall storyboard that encompasses the various elements of NG-CDI research projects to show how risk and governance maps across them and highlight the interconnections between them.

Philip is an Associate Professor at the Cambridge Judge Business School, University of Cambridge, and Director of the Centre for International Human Resource Management. He has a long-standing interest in technology, culture, and leadership and has worked extensively with a number of public sector, private sector and third sector organisaitions, and works with boards of directors looking at issues of transformations.

Automation in 5G Network Slicing: Challenges on resource control and service assurances
Professor Ning Wang, University of Surrey, 5G/6G Innovation Centre. 10/05/2022

This talk focuses mainly on the technical challenges relating to network automations in supporting resource control and service assurances in the context of 5G network slicing. In particular, Ning discusses key strategies in performing resource optimizations within/across heterogeneous 5G network segments (RAN, Transport and Core) and how they can be coordinated through end-to-end orchestration. Requirements on the end-to-end service assurance in dynamic network conditions are also elaborated.


​Ning received his PhD degree at University of Surrey in 2004 and he is currently a full-Professor at the Institute for Communication Systems (ICS) at the University of Surrey. His general research interests include 5G and beyond networking, Future Internet, network autonomics, multimedia content delivery, as well as Quality of Services (QoS) and Quality of Experiences (QoE).

Modern agent-based control for digital infrastructures
Marco Perez Hernandez, University of the West of England. 16/02/2022

Digital infrastructures are complex, elastic and inherently distributed. Various research efforts have addressed the use of agent-based systems in the control of critical infrastructures. One of the key promises of such systems is to enable distributed autonomous decision-making without compromising flexibility, scalability and portability. However, it has been only until the last decade that developments in virtualisation have enable real distribution of intelligence along the digital infrastructure and new challenges have emerged for the engineering of these control systems. In this talk we will review core concepts of the agent control systems and the current trends with application to control of modern digital infrastructures.

Marco is Lecturer of Computer Science at the University of the West of England (UWE) Bristol and Research Collaborator of the Distributed Information and Automation Lab (DIAL) of the University of Cambridge, and his research has addressed challenges related to the design and evaluation of agent-based supervisory control architectures for the network infrastructure.

Intent Based Network Testing & Monitoring
Charalampos Rotsos, Lancaster University. 16/03/2022

Autonomic network management can greatly improve service delivery times and reduce operational costs for telecommunication providers. Nonetheless, automation mechanisms are far from perfect, and they can occasionally have a negative effect on network functionality (sub-optimal or incorrect network configurations, delivery delays).

In this talk, we are going to discuss the challenges to validating and monitoring autonomic network management. Motivated by the design of our business intent layer, we are going to discuss how intent-based networking can be used not only to drive service deployment but can equally define precise network testing and monitoring goals. Furthermore, we are going to present the design of new tools, that can support automated network testing and monitoring. NEAT is a multi-layer network testing platform, providing a holistic environment to validate the correctness of a network configuration in realistic network conditions. Unimon is a novel service monitoring framework providing just-in-time lightweight service monitoring.

Charalampos is a Lecturer in Computer Networks and Networked Systems at Lancaster University. His research focus is in network service management and orchestration, network programmability and monitoring and cloud operating systems. 

BT Thought Leadership 2020/2021

A series of linked talks in 2020/2021 by principal academics on the NG-CDI project, hosted by the BT Thought Leadership programme...

Next Generation Converged Digital Infrastructure. 07/10/2020

Professor Nicholas Race.

Network Systems, University of Lancaster. Principal Investigator for NG-CDI.

The UK’s Digital Infrastructure is critical to the commercial and social activities and success of the country. It is essential that this infrastructure continues to be world leading. To keep ahead we need an infrastructure which responds quickly to changing needs, and at minimum cost. The creativity of the whole ecosystem will give rise to opportunities that we cannot predict. This means that services need to be configured in software rather than hardware to reduce the barriers to experimentation and scaling. Distributed autonomic technologies offer the opportunity to manage the expanding scale and complexity, and support faster ways to assess opportunities and risks, make decisions, and simplify service delivery. The operation of such an infrastructure will require new skills, cultures and practices. Nick introduces the research underway to deliver these aims & describes the approaches being taken and how the different aspect of the architecture fit together. 

Intent Based Networking. 03/11/2020

Professor Ning Wang, Networks, University of Surrey. 

Dr Harris Rotsos, Computer Networks, University of Lancaster.

Increasing the rate of delivery and value of new services will depend on smarter ways to capture customer needs and translate these into service definition and delivery. The research is investigating the capture of customer intents in machine-readable ways. The research covers not only service creation and DevOps, but also methods to maintain or re-negotiate service levels in real-time in the face of changing network dynamics, using autonomous distributed agent architectures.

Intelligent Asset Management for Service Assurance & Infrastructure Management. 08/12/2020

Dr Ajith Parlikad.

Asset Management, University of Cambridge, Institute for Manufacturing.

Adding intelligence to network assets offers the possibility that the infrastructure can trigger appropriate maintenance processes. Prognostic maintenance scheduling concentrates engineering effort on reducing the risk to customer service and costs.

Assets can learn from their own experiences, or from swarms of similar assets to anticipate their remaining useful life and co-operatively decide the best means to maintain service, reconfiguring themselves or calling for human help.

Risk models embrace the likely propagation of problems across the regions of the network and between other networks such as the power network. The prospect is offered of determining the best action at the time, based on the dynamics of the existing traffic pattern.

Network Assurance through Massive on-line Anomaly Detection. 13/01/2021

Prof. Idris Eckley.

Mathematics & Statistics, Lancaster University.

New statistical techniques have been developed which are able to monitor very large data streams and identify anomalies in near real time.

Distinguishing between normal randomness and specific types of anomalies is essential for determining when a pattern of events means an intervention is needed.

Multivariate techniques extend this to distinguish real system changes from spurious changes in the data.

Such techniques greatly increase the power and discrimination of Test and Diagnostic processes which deliver Network Assurance.

Technology, Risk & Organisations. 10/02/2021

Dr Philip Stiles.

Corporate Governance & Co-Director of the Centre for International Human Resources Management, Judge Business School, University of Cambridge.

New statistical techniques have been developed which are able to monitor very large data streams and identify anomalies in near real time.

Distinguishing between normal randomness and specific types of anomalies is essential for determining when a pattern of events means an intervention is needed.

Multivariate techniques extend this to distinguish real system changes from spurious changes in the data.

Such techniques greatly increase the power and discrimination of Test and Diagnostic processes which deliver Network Assurance.

World Models and Digital Networks. 09/03/2021

Prof. Rob Piechocki.

Wireless Systems, Turing Fellow, University of Bristol.

In this lecture I will review recent achievements in machine learning underpinned by learning representations in an un(self)supervised paradigm. Such techniques are at the heart of the latest and best performing language models (BERT, GPT-3), contrastive learning computer vision or protein folding predictors (AlphaFold2). The common feature of such techniques is an attempt to build a fundamental understanding of the model i.e. its “world model”, before a subsequent attempt is made to solve the given task. This is in stark contrast to the state-of-the-art techniques (including ML/AI) currently used in digital networks, where the algorithms are specifically crafted to solve the given task(s) and trained from the outset to achieve this. Can digital networks perform better by initially learning their own digital world models?  I will present the case in favour of this view, and not shy from listing arguments against it. 

Internal research presentations

A series of research presentations by NG-CDI Academics and PhD students...

Detecting Emergent Phenomena in Throughput Data Using FAST
Ed Austin, Lancaster University. 24/06/2021

Throughput data measures the volume of internet traffic passing through a point on the BT network. Over the course of each day the volume of throughput is expected to follow a similar shape, however deviations from this shape can occur and these may indicate that a fault or outage has occurred. In this talk I will present a FAST, a novel method for detecting these deviations in real time. Furthermore, I will present some results from applying FAST to BT data, and then propose extending the method to monitor multiple points on the BT network.

Ed Austin is currently a Research Associate working in Lancaster University on anomaly detection on the NG-CDI project, and a 2nd year PhD student completing a project titled “Novel Methods for the Detection of Emergent Phenomena Within Streaming Data”, supervised by Prof. Idris Eckley. 

Ed's research focusses on the detection of anomalies within streaming data, using techniques from online changepoint detection, nonparametric statistics, and functional data analysis to detect faults within telecommunications data in real time

Towards an Automated Testing Platform

Will Fantom & Paul Alcock, Lancaster University. 13/05/2021

As the network function virtualisation (NFV) paradigm evolves, the diversity of VNF entities available to operators expands. Technologies such as containers and unikernels pose new challenges to modern infrastructures. The NG-CDI project looks to bring an automation architecture to the deployment and management of these µVNFs, putting more control into the hands of developers through NetDevOps. The demonstration presented here shows how critical configuration errors can be caught prior to deployment within an emulation sandbox that represents the heterogeneous networks used today. Adding VM support the Mininet emulator, we show how µVNFs can be tested alongside userspace networking applications and SDN devices.

Will Fantom is a 3rd year PhD student at Lancaster University, investigating the roles that unikernels will play in next generation network infrastructure.

Paul Alcock is a 1st year PhD student with the networking group at Lancaster University, working in the domain of network service deployment, and researching how to create a testbed for virtual network function development with the popular network emulator mininet.

Predictive maintenance planning optimisation for telecommunication networks.

Alena Puchkova, University of Cambridge. 13/05/2021

The first attempts into developing an optimisation model for maintenance planning from a network point of view will be presented. Given predicted failure times of equipment across the whole network, the optimisation model aims to identify a network-wide optimal plan to perform predictive maintenance in such a way that the impact on the overall network performance and costs associated with maintenance are minimised. In addition, the model also identifies the best way to reroute traffic when nodes/links are shut down for maintenance or failed. To demonstrate the solution, a visualisation model has been developed which will be shown using several simple scenarios.

Dr Alena Puchkova is a Senior Research Associate in the Distributed Information & Automation Laboratory, Institute for Manufacturing, University of Cambridge, and she is looking into developing optimisation models to improve traffic rerouting and maintenance planning for assets across the network.

Towards a µvnf monitoring architecture - Ver.

Will Fantom, Lancaster University. 22/10/2020

µVNFs are the next step towards network softwarization. Taking the form of containers and unikernels, they offer to networking many parallels with microservice architectures. This presentation introduces UniMon, a monitoring approach that integrates monitoring directly into unikernel VNF pipelines. UniMon can provide detailed, contextual, timely data, yet remain in keeping with the minimal nature of the unikernel. Also introduced is UniProbe, a unikernel based monitoring architecture that inserts unikernel µVNFs into heterogenous network infrastructures to perform in-network monitoring, enabling decentralized monitoring and policy enforcement. Collecting application specific metrics, the UniProbe architecture can also couple with modern DevOps workflows, supporting greater network automation.

Will Fantom is a PhD student at Lancaster University, investigating the roles that unikernels will play in next generation network infrastructure.

P4Runtime-controlled UPF.

David Lake, University of Surrey. 22/10/2020

The talk will consider the development of P4 and P4 Runtime explaining how the notion of programmable networks has moved from the early days of OpenFlow. P4 as a programmable data-plane will be discussed as will P4Runtime as the network programming language associated with “Next-Generation SDN.” The Trellis architecture based around P4, P4Runtime and Stratum running on both hardware and software switches will be explained. Detail of current work to create a high-performance UPF based on P4 and P4Runtime using the open-source Free5GC components will explain how a mixture of hardware and software can provide the benefits of softwarization without compromising throughput in a 40Gbit/s UPF.

David Lake is a part-time PhD student at the University of Surrey and a member of the technical staff in the Office of the CTO at DELL.

IBN Demo presentation.

Mehdi Bezahaf, University of Lancaster. 24/09/2020

Recently, intent and Intent-Based networking (IBN) concepts have created enormous interest in academia and industry. The idea behind these concepts is to allow the user and the operator to express their intentions (i.e., a desire state or behavior) without the need to specify every technical detail of the process and operations to achieve it. In this talk, we introduce our technical demo on IBN from an Internet Service Provider (ISP) point of view. We present our intent implementation and demonstrate its utility in action in three use-cases (traffic engineering, predictive maintenance, and service protection period). It is still a work in progress and we keep updating our implementation with new use-cases.

Mehdi Bezahaf is a Senior Research Associate at Lancaster University, investigating future network architecture and network automation."

A series of short technical talks by Researchers and PhD students...

Tech Talks

Advanced Topics in Deep Learning
Xiaoyang Wang, Jonathan Thomas, University of Bristol. 08/02/2022

The last decades have witnessed rapid progress in Artificial Intelligence (AI), Machine Learning (ML) and Deep Learning (DL) technologies. As a subset of ML, DL is essentially using neural networks with multiple layers to solve ML problems. Through learning from large amounts of data, DL automates feature extraction, enabling predictions with incredible accuracy. In this talk, the basics of machine learning and deep learning will be introduced, followed by more details around advanced DL topics including deep reinforcement learning (DRL), self-supervised learning, graph neural networks and several specific applications in telecommunications.

Xiaoyang is a Research Associate in the Communication Systems and Networks Group, Department of Electrical and Electronic Engineering, University of Bristol. Her current research focuses on machine learning, especially reinforcement learning, and its applications in next-generation network management.

Jonathan is a PhD student in the Communication Systems and Networks Group, Department of Electrical and Electronic Engineering, University of Bristol. His research focuses on the application of cooperative Multi-Agent Reinforcement Learning within next-generation communications networks.

Bad Apples and Good Labels: Learning in Real-time Fault Detection
James Grant, Lancaster University. 18/01/2022

When monitoring telecommunications networks, it can be a relatively low-cost activity to flag anomalous patterns in data streams. It can be much more costly to ask engineers to respond to or counterfactually label these anomalies, and there is often is a high proportion of patterns which are anomalous in a statistical sense, but innocuous in an operational sense. This talk describes an approach we have developed which can work as a mediator between anomaly detection algorithms and experts. The approach, based on the so-called ‘apple tasting model’,  learns (in an online fashion, from limited expert feedback) to escalate only operationally-relevant anomalies to the expert, further automating the process of fault detection and monitoring.

James is a lecturer in Statistics at Lancaster University and a researcher on the NG-CDI project. His research considers the application of statistical thinking to sequential decision-making, particularly in settings with complex data structures. He has research interests in multi-armed bandits, recommender systems, and time series.

Unikraft: Fast, Specialized Unikernels the Easy Way
Alex Young, Lancaster University. 30/11/2021

The increasing softwarization of network technologies has mold computation as an essential resource to deliver next-generation network services. Nonetheless, the adoption of cloud technologies in order to manage and virtualize compute resources in the network infrastructure, limits the ability of operators to improve the resource efficiency and responsiveness of network services. In this talk, we will present a novel toolchain and library operating system for the cloud; Unikraft. Unikraft builds highly specialized unikernels, software bundles that consist of a target application along with just the operating system primitives and libraries features it needs to run. The Unikraft platform supports a wide range of VNF appliances (e.g. Click programs, HAproxy) and Unikraft kernels can run with minimal resource requirements (image size ~ 1Mb, memory footprint ~ 10 Mb) and fast boot times (30-40 msec). Furthermore, we will elaborate on the benefits of OS configuration tuning on cloud application performance and present Wayfinder, an holistic OS measurement platform which allows users to explore the configuration space of modern OSes and run custom OS benchmarking scenario in a reproducible way.


Alex is a 3rd year PhD student at Lancaster University focused primarily on the optimization of serverless applications and lightweight Virtual Network Functions (VNFs) constructed using specialized library Operating Systems known as Unikernels.

EPSRC Prosperity Partnerships

EPSRC has a strong track record of working closely with universities and business to develop high-quality funding programmes which deliver world-leading academic research whilst also delivering impact to business and the wider economy, through a variety of different routes.

Next Generation Converged Digital Infrastructure.

University of Bristol, University of Cambridge, Lancaster University, University of Surrey, BT. 01/11/2017

This EPSRC (Engineering and Physical Sciences Research Council) programme brings together a team of internationally renowned scientists and engineers working across top universities and BT. Together they will deliver the next generation converged digital infrastructure for the UK. This will create an agile, resilient network capable of meeting the future needs of our rapidly changing society and ensure that the UK’s digital infrastructure continues to be world leading. The partnership builds on long-term research collaborations between BT and each of the consortium’s members. This unique, multidisciplinary partnership with BT pioneers the way in which data science is harnessed with the latest research in networking.

Prosperity Partnerships: the first year.

EPSRC. 25/09/2018

New ideas and innovations are built on strong foundations of excellent research and fundamental science. That’s why, a year ago, EPSRC announced a really substantial investment to strengthen the links between the UK’s research base and business partners, driving innovation, growth and productivity.  Here’s a quick update on some of the partnerships a year on.

Kedar Pandya, Associate Director Business and User Engagement EPSRC

Bristol (web).png
Cambridge (web).png
Lancaster (web).png
Surrey (web).png
UKRI (web).png
BT 2 (web).png