Tutorial 1 Tutorial 1 Network Slicing: A holistic architectural approach, orchestration and management with applicability in mobile / fixed networks and clouds Alex Galis (UCL, UK) This half-day tutorial will present the concepts, design and technologies enabling Network Slicing (NS). NS is both an old and new idea. While its scope and definition continue to evolve, there is reasonable level common understanding of the key central role it will play in future network and service delivery scenarios. The significant value of network slicing architecture will be seen in terms of dynamic and non-disruptive control, operation and scaling of services in a network slice by the tenant of the slice. In order to achieve this, a fundamental change in how we administer networks is required in terms of abstraction, separation and mapping of forwarding, control and management aspects of a service. There exists a possibility of applying already mature methodologies such as SDN, NFV, service chains, network virtualization, etc. The challenge lies enabling an operator to flexibly use such tools in a light-weight manner with in a viable architectural framework. The first part of this tutorial presents the main slicing concepts and their history, key slicing characteristics and usage scenarios, Standard Organization activities in network slicing (ITU-T, IETF, ETSI, other SDOs), Slice Life Cycle Management, Management & Orchestration frameworks, Research & Development projects on Slicing. This part will conclude with identifying research and development challenges in Network Slicing. The second part presents industry use cases as a perspective on an overall architecture of network slicing and corresponding operations to demonstrate how an end-to-end slice couple be deployed and managed. It shows many existing standard techniques can be used to simplify network slice workflows. This part will wrap up with a report on progress and approach currently used for standardization of requirements of network slices. Monday, 5 November, 2018 09:00-12:30 2018-11-05T09:00:00+01:00 2018-11-05T12:30:00+01:00 Sala Bisogno Mini-conference Keynote Session 1 A Manageability Interface for Network, Telco and Industrial IoT John Leung (Intel Corporation, USA) Redfish is a simple, modern and secure manageability interface for cloud and web-based infrastructures. Redfish was released with a computer system model. Soon after, the Storage Network Industry Association released models for managing storage services. More recently, the Green Grid provided domain expertise which resulted in the release of models for manage data center facilities equipment. Over the last year, standards bodies have formed alliance partnerships with the DMTF to utilize Redfish for manageability in the Telecom and Industrial IoT domains. This presentation will provide an overview of Redfish and the status of efforts to extend Redfish with models to manage the network, telecom and Industrial IoT devices. 9:00-10:30 2018-11-06T11:00:00+01:00 2018-11-06T11:45:00+01:00 Sala Convegni MC1 Mini-conference Session 1 Edge and cloud computing 11:00-12:30 2018-11-05T11:00:00+01:00 2018-11-05T12:30:00+01:00 Sala Marconi Edmundo Madeira (UNICAMP) 11:00 11:18 1570467766 MC1.1 Adaptive Service Deployment using In-Network Mediation 14th International Conference on Network and Service Management 2018 - Full papers Serendipitous peer discovery is important for emerging Internet applications, particularly in dynamic environments (e.g., the IoT, ubiquitous and fog domains) where a large number of resources operate different services in any one locality and resource availability varies unpredictably over time. The current approach is to select services at design time based on offered providers and their reputation. This obviously has its limitations, particularly in terms of scalability and adaptivity, let alone the challenges of crossing vendor and operator divides. This work demonstrates how an application is better able to dynamically adapt to unforeseen environmental changes through in-network mediation of service requests. In our model, application developers express their service needs using intents. These are mapped to appropriate service providers with explicit consideration of the intermediate network. We design a general architecture and associated algorithms to realise intent formulation and processing for mapping application intents to service providers. Our results demonstrate the feasibility of adopting in-network mediation to enable adaptive application deployment using declarative intents. Abdessalam Elhabbash 1549305 Lancaster University United Kingdom (Great Britain) 0 Gordon Blair 105548 Lancaster University United Kingdom (Great Britain) 0 Gareth Tyson 830909 Queen Mary, University of London United Kingdom (Great Britain) 0 Yehia Elkhatib 351087 Lancaster University United Kingdom (Great Britain) 0 11:18 11:36 1570467774 MC1.2 Potential Traffic Savings by Leveraging Proximity of Communication Groups in Mobile Messaging 14th International Conference on Network and Service Management 2018 - Full papers Communication groups in social Internet applications, such as mobile messaging applications (MMAs), multiply the data transmissions, because every message has to be delivered to all members of the communication group. Thereby, they put a high load on mobile networks. As the number of recipients is still comparably small, the data-intensive user-generated content, such as self-recorded images or videos, cannot be handled efficiently in large, cloud-based content delivery networks. However, small communication groups, such as groups of friends or teams, might often be in close proximity, which can be leveraged to locally deliver messages by applying edge caching or device-to-device (D2D) communication. In this work, a simulation study is conducted to investigate these potential traffic savings in the mobile network. It is based on a realistic communication model of the MMA WhatsApp and utilizes different models for human mobility. The user mobility and MMA communication are simulated for a single day in a small city to obtain the ratio of messages, which could be potentially transmitted locally when utilizing edge caching and D2D communication. Michael Seufert 857079 AIT Austrian Institute of Technology GmbH Austria 0 Anika Schwind 1493916 University of Wuerzburg Germany 1 Marco Waigand 1597280 University of Würzburg Germany 0 Tobias Hoßfeld 118152 University of Würzburg Germany 0 11:36 11:54 1570467799 MC1.3 Service Placement for Hybrid Clouds Environments based on Realistic Network Measurements 14th International Conference on Network and Service Management 2018 - Full papers The ever increasing sophistication in IT services demands to consider dynamic and large-scale deployments in distributed environments built on top of federated public and private Clouds. Unfortunately, in these complex hybrid Cloud environments, it is extremely difficult to estimate the impact of even simple reconfigurations of the IT service architecture before enacting them. There is the need for new service management tools that are capable of exploring alternative IT service architectures, of accurately evaluating them to find out the most convenient one, and of reconfiguring the IT service accordingly. This paper presents Business-Driven Management as a Service Plus (BDMaaS+), that significantly evolves our previous work to consider the business-driven evaluation of IT services in hybrid Cloud environment, advanced IT service models that complex multi-tier workflows, and realistic network models. The experimental evaluation of BDMaaS+ to optimize a realistic large-scale IT service demonstrates that our tool is capable of finding a more convenient configuration that significantly reduce the total daily costs while granting at the same time agreed service levels. Mauro Tortonesi 105481 University of Ferrara Italy 1 Luca Foschini 273533 University of Bologna Italy 0 Genady Ya. Grabarnik 861547 St. John's University USA 0 Walter Cerroni 4986 University of Bologna Italy 0 Larisa Shwartz 134381 IBM Research USA 0 11:54 12:12 1570467862 MC1.4 Tackling Virtual Infrastructure Allocation in Cloud Data Centers: a GPU-Accelerated Framework 14th International Conference on Network and Service Management 2018 - Full papers Allocating IT resources to Virtual Infrastructures (VIs) (i.e., groups of VMs, virtual switches, and their network interconnections) is a problem which belongs to a class known to be NP-hard. Various approaches designed to run on CPUs have been proposed for reducing the search space and finding suitable allocation solutions. Most algorithms, however, face scalability issues when considering current cloud data centers comprising thousands of servers. To overcome these limitations, this work offers a set of allocation algorithms designed for Graphic Processing Units (GPUs). Experimental results evaluate the scalability of the algorithms and demonstrate their ability to handle three large-scale data center topologies. Lucas Nesi 1597378 Santa Catarina State University Brazil 0 Mauricio Pillon 1386838 Universidade do Estado de Santa Catarina & UDESC Brazil 0 Marcos Dias de Assunção 1211949 Inria France 0 Charles Christian Miers 309545 Santa Catarina State University Brazil 0 Guilherme Koslovski 310305 Santa Catarina State University Brazil 1 12:12 12:30 1570470693 MC1.5 lightMEC: A Vendor-Agnostic Platform for Multi-access Edge Computing 14th International Conference on Network and Service Management 2018 - Full papers Multi-access Edge Computing promises to improve mobile users experience by bringing storage and processing capabilities to the edges of the network. This approach allows third parties to deploy innovative services, like augmented reality, directly in the radio access network paving the way for new business models and new revenue stream for mobile network operators (MNOs). In this paper, we introduce lightMEC a lightweight solution for deploying mobile edge computing functionalities which allows hosting of low-latency and bandwidth-intensive applications at the network edge. Measurements conducted over a real-life test demonstrated that lightMEC can actually support practical MEC applications without requiring any change to the functionality of existing mobile network nodes both in the access and the core network segments. The significant benefits of adopting the proposed architecture are analyzed based on a proof-of-concept demonstration of the content caching use case. Tejas Subramanya 1438352 FBK CREATE-NET Italy 1 Giovanni Baggio 1600307 FBK CREATE-NET Italy 0 Roberto Riggio 146926 FBK CREATE-NET Italy 0 MC2 Mini-conference Session 2 Network Monitoring and Security 13:30-15:00 2018-11-05T13:30:00+01:00 2018-11-05T15:00:00+01:00 Sala Marconi Michael Seufert (AIT - Austrian Institute of Technology) 13:30 13:48 1570465447 MC2.1 Reproducing Popularity Dynamics of YouTube Videos 14th International Conference on Network and Service Management 2018 - Full papers To provide video streaming of user-generated contents (UGCs) with high quality and at low cost by maximizing the effect of CDN, CDN providers are required to adequately design CDN cache servers by accurately estimating the UGC view-count distribution. To achieve this goal in a practical time frame, we need to construct a simple time-series model that captures the transition of UGC popularity. Therefore, in this paper, we first analyze the daily view count (DVC) of YouTube videos over nine months and find that the DVC of YouTube videos obeys a lognormal distribution. As a simple time-series model of the DVC of each YouTube video, we propose the grouped MPP (gMPP), extending the multiplicative process (MPP) which is widely known as a simple time-series model generating a lognormal distribution. We also propose reproducing the DVC distribution of YouTube videos by using a superposed gMPP (SgMPP) that aggregating multiple gMPPs. The SgMPP can accurately reproduce the DVC distribution of YouTube videos with a low computational overhead, so we can expect to use the SgMPP as the input for computer simulations for designing various network components that require the popularity distribution of UGC, e.g., cache capacities. Noriaki Kamiyama 1077859 Fukuoka University Japan 0 Masayuki Murata 15627 Osaka University Japan 0 13:48 14:06 1570467036 MC2.2 Low False Alarm Ratio DDoS Detection for ms-scale Threat Mitigation 14th International Conference on Network and Service Management 2018 - Full papers The dynamically changing landscape of DDoS threats increases the demand for advanced security solutions. The rise of massive IoT botnets enables attackers to mount high-intensity short-duration "volatile ephemeral" attack waves in quick succession. Therefore the standard human-in-the-loop security center paradigm is becoming obsolete as the attackers can easily acquire such newly-fangled tools and botnets, high in sophistication and 'productivity'. To battle the new breed of volatile DDoS threats, the intrusion detection system (IDS) needs to also improve markedly, at least in reaction times and in automated response / mitigation. Designing such an IDS is a daunting task as network operators are traditionally reluctant to act --at any speed-- on potentially false alarms. The primary challenge of a low reaction time detection system is maintaining a consistently low false alarm rate. This paper aims to show how a practical FPGA-based DDoS detection and mitigation system can successfully address both of these challenges. Besides verifying the model and algorithms with real traffic "in the wild", we validate the low false alarm ratio. The key novel contributions of this paper are the 'false'-proof and the fast detection methods amenable to FPGA acceleration. As shown here, such methods can effectively mitigate the volatile ephemeral DDoS attacks and, accordingly are usable both in human out-of-loop and on-the-loop next-generation security solutions. Péter Orosz 1292951 Budapest University of Technology and Economics Hungary 0 Balázs Nagy 1514679 Budapest University of Technology and Economics & AITIA International Inc. Hungary 0 Pál Varga 283895 Budapest University of Technology and Economics Hungary 0 Mitchell Gusat 9291 IBM Zurich research laboratory Switzerland 0 14:06 14:24 1570467532 MC2.3 Passive Monitoring of HTTPS Service Use 14th International Conference on Network and Service Management 2018 - Full papers HTTPS is used today to secure the majority of web communications and so enhance user privacy. Therefore, traffic monitoring techniques must evolve to remain useful, especially to support security considerations, as for example detecting and filtering the forbidden uses of a web service. However, privacy should remain as intact as possible. This paper describes a new passive and transparent method to infer the use of a HTTPS service by extracting and interpreting only meaningful meta-data derived from the encrypted traffic without deeply profile individual users. We propose a model using the sizes of objects loaded in the HTTPS service as a signature, by leveraging kernel density estimation, supporting then a classification function. We assess this approach extensively on the Google Images Service although our approach remains valid for some other services. We succeed to achieve an accuracy of 99.18% when detecting particular keywords to be searched over a large dataset of 115,500 distinct keywords. Pierre-Olivier Brissaud 1544406 INRIA Nancy Grand Est & Thales France 1 Jérôme François 235420 INRIA Nancy Grand Est France 0 Isabelle Chrisment 128950 LORIA-TELECOM Nancy, Université de Lorraine France 0 Thibault Cholez 323457 LORIA / INRIA Nancy - Grand Est France 0 Olivier Bettan 1207449 Thales France 0 14:24 14:42 1570467816 MC2.4 Bitforest: a portable and efficient blockchain-based naming system 14th International Conference on Network and Service Management 2018 - Full papers Public key infrastructures (PKIs), or more generally secure naming systems, lie at the foundation of the security of any communication system. Without a trustworthy binding between user-facing names, such as domain names, and cryptographic identities, such as public keys, all security guarantees against active attackers come crashing down like a house of cards. Blockchains such as Bitcoin, by offering a decentralized yet secure public ledger, show promise as the root of trust for naming systems with no central trusted parties, greatly increasing their security compared to traditional centralized PKIs. Yet blockchain PKIs such as Namecoin and Blockstack tend to significantly sacrifice scalability and flexibility in pursuit of decentralization, hindering large-scale deployability on the Internet. We propose Bitforest, a secure naming system with an architecture combining a centralized yet only partially trusted name server with efficiently queryable verification data embedded in a novel data structure inside a cryptocurrency blockchain. Bitforest achieves decentralized trust and security as strong as existing blockchain-based naming systems while retaining most of the flexibility and performance of centralized PKIs, allowing fully-validating thin clients to look up and verify name bindings with comparable efficiency to traditional systems. We use both numerical simulation and real-world experiments to evaluate the performance of Bitforest compared with other naming systems, both centralized and blockchain-based, showing that its performance goals are indeed achieved. Yuhao Dong 920153 University of Waterloo Canada 1 Woojung Kim 1578468 University of Waterloo Canada 0 Raouf Boutaba 5035 University of Waterloo Canada 0 14:42 15:00 1570470336 MC2.5 In-Network QoE and KPI Monitoring of Mobile YouTube Traffic: Insights for Encrypted iOS Flows 14th International Conference on Network and Service Management 2018 - Full papers With the steady rise in OTT mobile video streaming traffic delivered using encryption protocols, network operators are faced with the challenge of monitoring their customers' Quality of Experience (QoE). Solutions for monitoring QoE-related KPIs are a necessary prerequisite to detecting potential impairments, identifying their root cause, and consequently invoking QoE-aware management actions. In this paper, we leverage a supervised machine learning approach to train QoE and KPI classifiers for mobile YouTube video streaming sessions using features extracted from encrypted QUIC traffic. With previous studies having shown different service behavior across different access networks (WiFi, mobile) and different OSs (Android, iOS), we go beyond related work and specifically address iOS measurements and models. We assess the performance of models trained on data from a lab WiFi environment and an iOS device through cross-validation, achieving promising results. Using the dataset collected in a lab WiFi network, and two additional datasets collected in operational mobile networks, we further report on the promising applicability of classifiers trained using the WiFi dataset when applied to traffic collected using mobile network probes. The implications of such findings show the potential to use the same classifiers for multiple usage scenarios, thus reducing efforts needed for data collection and training. Finally, we discuss the extent to which models previously trained for Android usage scenarios are applicable for the iOS platform. Irena Orsolic 1423824 University of Zagreb, Faculty of Electrical Engineering and Computing Croatia 1 Petra Rebernjak 1599869 University of Zagreb, Faculty of Electrical Engineering and Computing Croatia 0 Mirko Suznjevic 355524 University of Zagreb, Faculty of Electrical Engineering and Computing Croatia 0 Lea Skorin-Kapov 107654 University of Zagreb, Faculty of Electrical Engineering and Computing Croatia 0 Tutorial 2 Tutorial 2 Network Function Virtualization (NFV) Enabling Technology for 5G Faqir Zarrar Yousaf Communication networks' architecture and technologies are undergoing through their next evolutionary phase towards 5G. This evolutional phase is markedly different from the previous phases in that the complete concept of networking is undergoing a paradigm shift. Networks are now moving beyond as mere bit-pipes towards a highly agile and programmable entity that can be tailored to the varying demands of the various industry verticals while ensuring conformance to stringent performance requirements and expectations of 5G. New technologies and architectures are being developed that will provide the customers their own isolated network "slices" over the same infrastructure thus creating new use cases and business models. NFV is now recognized as one of the key enabling technologies for the management of 5G networks, and the NFV Management and Orchestration (MANO) framework and its various aspects is being specified keeping in view of management and orchestration of virtualized network functions and services. The goal of this tutorial is: 1. To highlight the challenges and requirements for the management of 5G networks; 2. To impart in-depth knowledge about the ETSI NFV MANO system framework; 3. To explore the features/capabilities/functions/services of NFV MANO system in view of the 5G networks requirements/challenges. This tutorial will also discuss the various integration options of the NFV MANO with the SDN networks and how the two technologies complement each other in imparting a truly programmable and agile network platform. At the end of the proposed 180 minute tutorial, the audience is expected to have an effective knowledge of NFV MANO system and its key role in the context of 5G. 13:30-17:00 2018-11-05T13:30:00+01:00 2018-11-05T17:00:00+01:00 Sala Bisogno MC3 Mini-conference Session 3 SDN and NFV 15:30-17:00 2018-11-05T15:30:00+01:00 2018-11-05T17:00:00+01:00 Sala Marconi Abbas Bradai (University of Poitiers) 15:30 15:48 1570465970 MC3.1 Understand Your Chains and Keep Your Deadlines: Introducing Time-constrained Profiling for NFV 14th International Conference on Network and Service Management 2018 - Full papers Fully-automated resource dimensioning is one of the key requirements for agile, DevOps-enabled network function virtualization (NFV) scenarios in which new service versions are continuously delivered and deployed to production. To enable and support these dimensioning processes, a priori knowledge about the performance behavior of the deployed service function chains (SFC) is collected using profiling solutions that automatically generate so-called SFC performance profiles. A challenge in those profiling processes is the huge configuration space of typical SFCs that need to be explored to collect enough information to accurately describe the performance behavior of the profiled SFC. In this paper, we introduce the concept of time-constrained profiling (T-CP) which profiles only a small subset of all possible SFC configurations and uses machine learning techniques to predict performance values for the remaining configurations to generate a full SFC performance profile within a given time limit. Using our novel, open-source T-CP prototype, we analyze the accuracy of the generated profiles using different selection algorithms to find good configuration subsets. We base parts of this analysis on real-world SFC performance measurements, which we make publicly available as open dataset. Manuel Peuster 1373829 Paderborn University Germany 1 Holger Karl 3972 Paderborn University Germany 0 15:48 16:06 1570467193 MC3.2 Load balancing and flow management under user mobility in heterogeneous wireless networks 14th International Conference on Network and Service Management 2018 - Full papers The utilization and size of today's wireless networks is continuously increasing, as more and more wireless communication technologies and connected devices are being added. As the use of multiple communication technologies is supported by modern devices, efforts have been made to allow these devices to utilize simultaneously and handover between different technologies. However, existing management frameworks and standards lack the intelligence to provide fine-grained network-wide optimizations. This despite the potential of dramatically increasing overall network performance (e.g., throughput) and user experience. To this extent, we present a multi-technology load balancing approach that can manage devices and steer traffic across different wireless technologies, in order to maximize the global throughput. This dynamic approach can be deployed on top of existing solutions and takes into account the specific characteristics of wireless networks and the mobility of stations. We present a mathematical problem formulation of load balancing traffic and devices across different wireless technologies. Furthermore we introduce a heuristic algorithm that ensures practical scalability. We demonstrate its ability to increase the network-wide throughput by more than 100% across a variety of scenarios and scalability up to 10000 devices. Tom De Schepper 1322649 University of Antwerp - imec Belgium 1 Steven Latré 1059359 University of Antwerp - imec Belgium 0 Jeroen Famaey 508641 University of Antwerp & Imec Belgium 0 16:06 16:24 1570470611 MC3.3 Consistent SDN Rule Update with Reduced Number of Scheduling Rounds 14th International Conference on Network and Service Management 2018 - Full papers Consistent operation of software-defined network (SDN) switches during the transient periods of forwarding rule updates is a critical issue. This paper studies the problem of updating SDN rules, while preserving two essential security and performance consistency properties: (1) Waypoint Enforcement which mandates that all packets traverse a specific checkpoint (e.g., firewall), and (2) Loop-Freedom that prevents forwarding packets along a loop. To guarantee these properties, we schedule rule updates in multiple rounds. To reduce the time that the network stays in the transient period of updating the switches, we have to solve the NP-hard problem of minimizing the number of update rounds. To this end, we design a fast algorithm called RRS which can be applied to very large networks. Our experiments on a large dataset of 28K scenarios show that RRS achieves a 323x improvement in the median of execution time compared to solving the exact Mixed Integer Program (MIP) formulation. Mahdi Dolati 1518176 Iran & University of Tehran Iran 0 Ahmad Khonsari 152276 University of Tehran Iran 0 Majid Ghaderi 253929 University of Calgary Canada 0 16:24 16:42 1570470873 MC3.4 Proactive Access Point Driven Handovers in IEEE 802.11 Networks 14th International Conference on Network and Service Management 2018 - Full papers In large and densely deployed IEEE 802.11 networks, a fast and seamless handover scheme is an important aspect in order to provide reliable connectivity for mobile users. However, IEEE 802.11 only supports decentralized, reactive and mobile node-driven handovers resulting in long handover times, packet loss due to interrupted connectivity and sub-optimal access point selection. Recently, centralized approaches have been developed that try to solve many of the challenges but these are mostly proprietary, reactive and require changes to mobile node stacks. In this paper, we propose a novel handover solution based on the principle of Software Defined Networking, that addresses the aforementioned challenges. Using virtualization and softwarization, we shift the traditional mobile node-driven handovers to the access point, while maintaining compliance with legacy devices. Moreover, we develop a proactive handover algorithm ADNA, which combines network state, traffic load and node mobility information. We evaluate our approach extensively in a testbed, showing that it outperforms existing approaches by improving the overall throughput by 116% while reducing the number of handovers by 44% on average. Ensar Zeljković 1477722 University of Antwerp - imec & IDLab Belgium 1 Steven Latré 1059359 University of Antwerp - imec Belgium 0 Johann M. Marquez-Barja 261988 University of Antwerpen - imec & IDLab Research Group - imec Belgium 0 Andreas J. Kassler 111482 Karlstad University Sweden 0 Roberto Riggio 146926 FBK CREATE-NET Italy 0 16:42 17:00 1570474657 MC3.5 Revive: A Reliable Software Defined Data Plane Failure Recovery Scheme 14th International Conference on Network and Service Management 2018 - Full papers In Software-Defined Networking (SDN) links and switches (nodes) from data plane suffer from failure and impact network operations. In the presence of such failures, switches can use reactive or proactive recovery scheme. In the reactive scheme, switches contact the controller after detecting a link failure to get a backup route setup; whereas, switches locally redirect the traffic to the backup route without controller's intervention in the case of the proactive scheme. In this paper, we propose a hybrid recovery scheme, called Revive, where the controller proactively installs backup routes only in a subset of the switches between a source-destination pair. In addition, we judicially configure the routes in Revive to meet the application and the reliability demand while efficiently utilize the network resources. Extensive experimental results in Mininet using real topologies illustrate the benefits of Revive compared to its counterparts. Israat Haque 145146 Dalhousie University Canada 1 Ma Moyeen 1598350 Dalhousie University Canada 0 Keynote Session 1 Segment Routing: Solution, Use-case, Deployment and research opportunity Clarence Filsfils (CISCO, Belgium) Segment Routing (SR) is a modern variant of source routing, which is being developed within the SPRING and IPv6 working groups of the IETF. Segment Routing gives the possibility to include a list of instructions (called segments) in the packet headers. This list of segments influences the forwarding path of the packets and can also provide instructions to be performed on a packet in a given node. End-to-end policies can be realized without creating any per-flow state in the network. The Segment Routing architecture seeks the right balance between distributed intelligence and centralized optimization. Tuesday, 6 November, 2018 9:00-9:45 2018-11-06T11:00:00+01:00 2018-11-06T11:45:00+01:00 Sala Convegni Keynote Session 2 Keynote Session 2 5G capabilities as Enablers for Automated Driving Johannes Springer (Deutsche Telekom AG / T-Systems International, Germany) 9:45-10:30 2018-11-06T11:45:00+01:00 2018-11-06T12:30:00+01:00 Sala Convegni TS1 Technical Session 1 Monitoring and Quality of Service 11:00-12:30 2018-11-06T09:00:00+01:00 2018-11-06T10:30:00+01:00 Sala Convegni Marc-Oliver Pahl (TUM) 09:00 09:30 1570470109 TS1.1 Network Performance Monitoring with Flexible Models of Multi-Point Passive Measurements 14th International Conference on Network and Service Management 2018 - Full papers Many management actions for networking infrastructures require to simultaneously consider the state of several network elements. This is particularly critical in the case of reconfigurable deployments, such as Virtual or Software-Defined Networks, to scale the affected equipment up and prevent performance bottlenecks. In this light, we present dPRISMA (distributed Passive Retrieval of Information, and Statistical Multipoint Analysis), a passive monitoring system intended to fit statistical models for network measurements and raise alarms in the case of extreme behaviors. As distinguishing features, dPRISMA relies on cost-effective multi-point network measurements, and is able to select a suitable parametric model optimizing the trade-off between fitting and complexity. Therefore, it can (i) correlate records collected from several vantage points and detect where performance issues are most likely to appear; (ii) adjust alarms in terms of the probability of events; and (iii) adapt its behavior to dynamic network conditions while presenting a fair identification of anomalous situations. We evaluate dPRISMA with experiments both in virtual environments and with real-world data to provide evidences of its applicability. Daniel Perdices 1599680 Naudit HPCN, SL Spain 0 David Muelas 1294161 Universidad Autónoma de Madrid Spain 0 Luis de Pedro 213373 Universidad Autonoma de Madrid Spain 0 Jorge E. López de Vergara 130998 Universidad Autónoma de Madrid (UAM) & Naudit High Performance Computing and Networking Spain 0 09:30 10:00 1570470751 TS1.2 INTCollector: A High-performance Collector for In-band Network Telemetry 14th International Conference on Network and Service Management 2018 - Full papers In Software-Defined Networking (SDN), monitoring is an essential function to provide information about the network and help the SDN controller to make network controlling decisions. In-band Network Telemetry (INT) is a new method that can provide real-time, fine-grained, and end-to-end network monitoring. INT works by embedding network information into every packet. At the final switch in the path, INT extracts network information into report packets and sends the reports to a collector. However, the huge amount of data from INT requires high processing capability of the collector. We present the design and implementation of INTCollector, a high performance collector for INT. We propose a mechanism to extract important network information, called event, from INT raw data. The mechanism filters network events, reduces the number of metric values that need to be stored, reduces CPU usage and storage cost while still ensuring to capture and store all important network information. INTCollector has two processing flows: a fast path to process INT report packets, and a normal path to process events and store metric values into a database. The fast path is accelerated by eXpress Data Path (XDP) - a Linux in-kernel fast packet processing framework. Our calculation shows that event detection can massively reduce the amount of data need to be stored (two to three orders of magnitude in our test scenario). The evaluation shows that INTCollector can process INT reports at the rate of 1.2 Mpps with 8\% of CPU when running with software NIC. We expect better result can be achieved with XDP supported hardware NICs. Tu Van Nguyen 1335095 POSTECH Korea 0 Jonghwan Hyun 868609 POSTECH Korea 0 James W. Hong 2797 POSTECH Korea 0 10:00 10:30 1570473370 TS1.3 A Fair Share for All: Novel Adaptation Logic for QoE Fairness of HTTP Adaptive Video Streaming 14th International Conference on Network and Service Management 2018 - Full papers This paper presents a novel adaptation logic for HTTP adaptive streaming (HAS), which achieves not only a high Quality of Experience (QoE) but also high QoE fairness among independent and heterogeneous clients. The algorithm forces video clients to adapt the requested quality level based on the current network conditions and their individual bit rate requirements, such that the overall quality levels selected by all currently active streaming clients are fairly distributed, i.e., they do not diverge too much. The design of the algorithm is inspired by the well-known Transmission Control Protocol (TCP) congestion control, and drives heterogeneous clients to independently converge on similar quality levels without the need for communicating with each other and/or with a centralized controller in the network. By defining quality levels with equal visual quality, and preparing video representations accordingly, the quality level fairness is extended to QoE fairness. In this work, the design of the algorithm is described and a simulative performance evaluation is conducted to compare the QoE and QoE fairness of the proposed algorithm with other HAS adaptation logics. Michael Seufert 857079 AIT Austrian Institute of Technology GmbH Austria 0 Nikolas Wehner 1377215 AIT Austrian Institute of Technology GmbH Austria 0 Pedro Casas 244435 Austrian Institute of Technology (AIT) Austria 0 Florian Wamser 720551 University of Wuerzburg Germany 0 TS2 Technical Session 2 Machine Learning 13:30-15:00 2018-11-06T13:30:00+01:00 2018-11-06T15:00:00+01:00 Sala Convegni Yehia Elkhatib (Lancaster University) 13:30 14:00 1570467720 TS2.1 A Machine Learning Approach for IEEE 802.11 Channel Allocation 14th International Conference on Network and Service Management 2018 - Full papers Today's communication is mainly done over wireless networks, with IEEE 802.11 (Wi-Fi) at the forefront. There are billions of devices and millions of access points (APs), but only very few non-overlapping channels. As a result, the performance of Wi-Fi devices is severely degraded, because perfect channel allocation - with every AP alone in its channel - is close to impossible. Even in situations where all networks are under centralised control, existing approaches quickly tend to be either unscalable or suboptimal. By focusing on a subset of problems, identifying Wireless Local Area Networks (WLANs) that severely interfere with each other, performance can be improved even in such a complex situation. We tackle this problem through machine learning and coin it Bad Neighbour Detection (BND). Based on this output alongside monitoring data about the networks' activity, we then propose a channel allocation that optimises performance and as a side effect, stabilises networks that we do not control. We evaluate our approach in a field trial and show that we significantly improve the experience for users, eliminating virtually all interference-related issues. Olivier Jeunen 1597217 University of Antwerp Belgium 0 Patrick Bosch 1308619 University of Antwerp - imec Belgium 1 Michiel Van Herwegen 1597222 Technicolor Belgium 0 Karel Van Doorselaer 1145145 Technicolor Belgium 0 Nick Godman 1597226 Technicolor Belgium 0 Steven Latré 1059359 University of Antwerp - imec Belgium 0 14:00 14:30 1570470758 TS2.2 Cross-Network Behavioral Clustering for Managed Security Service Providers 14th International Conference on Network and Service Management 2018 - Full papers Managed Security Service Providers (MSSP) oversee and protect customer networks often varying in the level of pre-installed security defense appliances and capabilities. This imbalance makes it challenging to offer a consistent level of service across clients. This paper presents an approach for raising the level of defense by indirectly utilizing the threat detection capabilities of secure networks based on behavioral similarity among cross-network entities. Initially, we present a holistic architectural view of the components we have deployed in order to efficiently ingest, process and analyze massive amounts of raw data from various perimeter network devices of MSSP customers. Our data analysis approach is based on entities clustering and a risk estimation routine leveraging on noisy labeling derived from advanced security appliances. Specifically, entities are clustered according to a list of aggregated metrics, characterizing the communication between local and remote network devices. Finally, we present the rationale behind the adoption of specific clustering approaches, as well as an optimization routine implemented to appropriately select the free parameters of the clustering sub-process. Ilias Kotinas 1600364 PCCW Global Greece 0 Georgios Gkroumas 1599415 PCCW Global Greece 0 Kostas Giotis 1600365 PCCW Global Greece 0 Theocharis Tsigkritis 1600366 PCCW Global Greece 0 Paris Mermigkas 1600368 PCCW Global Greece 0 14:30 15:00 1570467848 TS2.3 Predicting Distributions of Service Metrics using Neural Networks 14th International Conference on Network and Service Management 2018 - Full papers We predict the conditional distributions of service metrics, such as response time or frame rate, from infrastructure measurements in a cloud environment. From such distributions, key statistics of the service metrics, including mean, variance, or percentiles can be computed, which is often needed for predicting SLA conformance or enabling service assurance. We model the distributions as Gaussian mixtures, whose parameters we predict using mixture density networks, a class of neural networks. We apply the method to a video-on-demand service and a KV store running on our lab testbed under different load patterns. The results validate the suitability of the method suggest that accurate prediction can be made. In the case of predicting the mean of frame rate or response time, the accuracy matches that of random forest, a baseline model. Forough Sahab Samani 1597364 KTH Royal Institute of Technology Sweden 0 Rolf Stadler 149700 KTH Royal Institute of Technology Sweden 0 PS1 Poster Session 1 15:30-17:00 2018-11-06T15:30:00+01:00 2018-11-06T17:00:00+01:00 Sala Convegni 15:30 15:30 1570467442 PS1.1 Latency Modelling in IEEE 802.11 Systems with non-IEEE 802.11 Interfering Source 14th International Conference on Network and Service Management 2018 - Full papers Centralised wireless network management becomes increasingly common but requires information about wire- less links to be able to properly use the existing spectrum. IEEE 802.11 employs the Distributed Coordination Function (DCF), which uses a decentralised Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) approach to control the access to the wireless medium. Both IEEE 802.11 and non- IEEE 802.11 sources can interfere with the transmission and degrade performance, but the impact of non-IEEE 802.11 in- terfering sources has not been extensively researched yet in literature. Only systems without a non-IEEE 802.11 interfering sources are considered in existing models. In this paper we explore the impact of non-IEEE 802.11 interference through realistic measurements in a large-scale wireless testbed. Based on these observations, we discuss the impact of an interfering source and identify the main components that affect the delay. We also propose an outline of a possible analytical model that we will extend in future work. Patrick Bosch 1308619 University of Antwerp - imec Belgium 1 Steven Latré 1059359 University of Antwerp - imec Belgium 0 Chris Blondia 1131621 imec - University of Antwerp Belgium 0 15:30 15:30 1570467520 PS1.2 ARCD: a Solution for Root Cause Diagnosis in Mobile Networks 14th International Conference on Network and Service Management 2018 - Full papers With the growth of cellular networks, the supervision and troubleshooting tasks have become troublesome. The present paper specifies a root cause diagnosis framework that identifies the major contributors (equipments, services, user groups) to the network overall inefficiency, classifies these major contributors into groups and explores the dependencies between the different groups. Our solution provides telecommunication experts with a graph summing up the fault locations and their eventual dependencies which helps them trigger the adequate maintenance operation. Maha Mdini 1596997 IMT Atlantique -- Exfo France 0 Gwendal Simon 205964 IMT Atlantique France 0 Alberto P Blanc 157357 IMT Atlantique France 0 Julien Lecoeuvre 1596998 Exfo France 0 15:30 15:30 1570467623 PS1.3 eDoS Mitigation for Autonomic Management on Multi-Tier IoT 14th International Conference on Network and Service Management 2018 - Full papers In this age of the Internet of Things and ubiquitous computing, autonomic management has become a critical component in cloud platforms. Autonomic management helps systems adapt seamlessly and efficiently to rapidly fluctuating workloads. However, economic Denial of Sustainability (eDoS) attacks can directly target the autonomic management to waste resources. In this paper, we propose an eDoS mitigation framework that incorporates online anomaly detection with our Elascale autonomic management system to thwart eDoS attacks in real-time. This allows the detection system to be application-agnostic as this framework utilizes only resource statistics of the monitoring applications. We present the design and implementation of our anomaly detection framework with Elascale. We evaluate Hierarchical Temporal Memory (HTM) and Tukey with Relative Entropy against spatial and temporal anomalies. Our results prove that the HTM-based anomaly detection method outperforms with significant accuracy. Rajsimman Ravichandiran 1594553 University of Toronto Canada 0 Hadi Bannazadeh 869349 University of Toronto Canada 0 Alberto Leon-Garcia 393083 University of Toronto Canada 0 15:30 15:30 1570467734 PS1.4 Towards Dynamic Fog Resource Provisioning for Smart City Applications 14th International Conference on Network and Service Management 2018 - Full papers Over the past few years, the Internet of Things (IoT) has transformed the Smart City concept into an attractive and relevant opportunity. Smart Cities aim to connect billions of objects of everyday life to the Internet to improve sustainability and citizen welfare. Fog Computing has been introduced to provide scalable and low latency services for the future IoT use cases by placing cloud resources at the edges of the network. Nevertheless, crucial challenges still remain in the Fog Computing domain. One of them is providing proper resource provisioning while reducing allocation costs, maximizing energy efficiency and minimizing latency. Therefore, in this paper, a Resource Discovery Service based on Peer-to-Peer (P2P) Distributed Hash Tables (DHT) is presented to enable automated resource discovery functionalities for Smart City services. The proposed approach provides a flexible way of exchanging resource allocation information between the Fog and the Cloud Layer. Evaluations have been conducted to demonstrate the performance of implementing information dissemination systems based on DHT protocols. Performance ratios higher than 99%, including latencies under 0.7 seconds for a static context have been obtained. Results show that DHTs provide scalable discovery solutions showing the full applicability of the proposed approach in the Smart City environment. José Santos 1500073 Ghent University - imec, IDLab Belgium 1 Tim Wauters 114405 Ghent University - iMinds Belgium 0 Bruno Volckaert 1503169 Ghent University-imec & IBBT Belgium 0 Filip De Turck 97039 Ghent University - imec Belgium 0 15:30 15:30 1570467758 PS1.5 Overcoming Network and Security Management Platform Gaps in Federated Software Networks 14th International Conference on Network and Service Management 2018 - Full papers Software networks are on the verge of replacing traditional communications network infrastructures on a large scale; they are composed of software-defined networks, virtualized network functions, and compute resources such as virtual machines. While software networks thrive on scalability, high dynamics, and flexibility, they also are inherently tied to conceptually new challenges for network management, i.e., the tasks carried out by network operators in order to provision, maintain, and optimize network-based IT services. While individual management tools are made available for each software network component by the vendors, integrated management architectures, which enable the common management of components across the heterogeneity of vendors and types as well as models, need to be redesigned to achieve the same level of maturity and usefulness compared to their counterparts in traditional networks. This paper intends to foster research on integrated management for software networks systematically by first defining the key characteristics of software networks and the current management tools. Based on federated software networks, which span over several distributed management domains, gaps in management functionality are identified and lead to the specification of requirements for modern integrated management platforms. We propose measures to close the identified gaps and derive open research issues as well as implementation recommendations. Michael Steinke 1593059 Bundeswehr University Munich Germany 1 Wolfgang F. Hommel 148179 Bundeswehr University Munich & Munich Network Management Team Germany 0 15:30 15:30 1570467788 PS1.6 XMPP as a scalable multi-tenants isolation solution for ONOS-based Software-Defined Cloud Networks 14th International Conference on Network and Service Management 2018 - Full papers BGP/MPLS IP VPNs has emerged as a stable solution for L3 tunneling and inter-sites routing and connectivity system over a Wide Area Network (WAN). This solution can be extended via the Extensible Messaging and Presence Protocol (XMPP) to provide simple and scalable multi-tenants isolation technology for cloud networks with Software Defined Networking (SDN) scheme. Also, it provides interoperability between BGP-based WANs and internal SDN-based cloud networking, and avoids inter-subnet or north-south traffic to be routed via central network gateway acting as single point of failure. In this work, we overview this solution and introduce the structure of our implementation of the XMPP as a South-Bound Interface (SBI) framework for the Open Networking Operating System (ONOS); a popular SDN controller Our work allows the deployment of new use-cases based on ONOS. Tomasz Osiński 1480518 Orange Polska Poland 0 Abdulhalim Dandoush 192104 ESME Sudria France 0 15:30 15:30 1570467793 PS1.7 Log Analysis via Space-time Pattern Matching 14th International Conference on Network and Service Management 2018 - Full papers This paper proposes a new methodology inspired from pattern matching and able to find alarm correlations with or without prior knowledge about the monitored system. The data structure can store every observed pattern of correlated alarms by processing logs online. It can be queried to extract the patterns of alarms leading to an arbitrary failure. First, we propose a framework able to represent alarm logs according to spatio-temporal dependencies. Second, we design a new scalable data structure, able to store every observed pattern of alarms, and validate it by simulation. Third, we show how to exploit this data structure for fault diagnosis. Anne Bouillard 1542417 Nokia Bell Labs France 0 Marc-Olivier Buob 235613 Nokia Bell Labs France 0 Maxime Raynal 1597308 Université de Grenoble France 0 Achille Salaün 1597310 Nokia Bell Labs France 1 15:30 15:30 1570470479 PS1.8 Outlier Detection for Distributed Services using Multi-Frequency Patterns 14th International Conference on Network and Service Management 2018 - Full papers Outlier detection has been commonly used in large scale distributed systems to detect abnormalities and service failures. Traditional abnormality detection methods are mostly based on statistical analysis or simple thresholds to detect outliers. However, those solutions do not consider the periodic patterns of data which is a common feature in many systems due to users' repetitive activities on a daily basis. In our work, we investigate a method that can significantly improve the effectiveness of outliers detection through utilizing periodic patterns in the system data. To this purpose we use State-Frequency-Memory (SFM) Recurrent Neural Networks (RNN) to analyze the cyclical patterns of monitoring data and use cumulative sum (CUMSUM) to identify outliers. Our real data evaluation shows that our detection method can achieve 99\% accuracy. Finally, our method also applies to many other fields, such as network traffic security monitoring, bank fraud detection, etc. Lu Xu 1599430 Beihang University P.R. China 0 Zhongzhi Luan 1600771 Beihang University P.R. China 0 Carol J Fung 150276 Virginia Commonwealth University USA 0 Guang Wei 1600774 Beihang University P.R. China 0 Depei Quian 214867 Beihang University P.R. China 0 15:30 15:30 1570470523 PS1.9 Blockchain-based Infrastructure Sharing in 5G Small Cell Networks 14th International Conference on Network and Service Management 2018 - Full papers Recently, the interest in using Blockchain as a secure and distributed ledger has increased dramatically. Although the main purpose of Blockchain by means of Bitcoin was about cryptocurrency and peer to peer transactions, its application to other systems has been widely used. One of the fields that has potential possibilities to benefit from BlockChain features is telecommunication. BlockChain can be applied in case of management of various networks to reduce some expenses. In this position paper, we apply a BlockChain network with smart contract in the cellular mobile networks. The Blockchain can provide a distributed HSS in a way that the core networks of different operators can use it in a secure manner. Moreover, the smart contract can act as a distributed Self Organizing Network features to handle self-transactions among mobile operators in return of sharing small cells' infrastructure. Babak Mafakheri 1569822 FBK CREATE-NET & University of Bologna Italy 0 Tejas Subramanya 1438352 FBK CREATE-NET Italy 0 Leonardo Goratti 1471036 Zodiac Aerospace Germany 0 Roberto Riggio 146926 FBK CREATE-NET Italy 0 15:30 15:30 1570470581 PS1.10 Mapping the allocation of resources for 5G slices to the k -MUTEX with n instances of m resources problem 14th International Conference on Network and Service Management 2018 - Full papers The introduction of slicing in 5G networks requires to consider the infrastructures as shared between not only users but also between independent services with heterogenous requirements. This problem can be seen as a variation of the k-mutex problem. In this paper we introduce three contributions. First we propose a formal model for the problem of allocation of resource for 5G slicing. Then we present adaptation of an existing algorithm for a more general k-mutex problem to our problem. The third and last contribution are the experimental results of tests run in our simulator RAMONES showing the performances compared to other algorithms. Guillaume Fraysse 1533352 Sorbonne Université & Orange Labs France 1 Jonathan Lejeune 1535205 Sorbonne Université & LIPN France 0 Julien Sopena 275046 Université Pierre et Marie Curie France 0 Pierre Sens 125381 Université de Paris 6 France 0 TS3 Technical Session 3 Security and Resiliency Wednesday, 7 November, 2018 09:00-10:30 2018-11-07T09:00:00+01:00 2018-11-07T10:30:00+01:00 Sala Convegni Jorge Lobo (ICREA/UPF) 09:00 09:30 1570467802 TS3.1 Combined Control and Data Plane Robustness of SDN Networks against Malicious Node Attacks 14th International Conference on Network and Service Management 2018 - Full papers In the context of software-defined networking (SDN), we address a variant of the controller placement problem (CPP), which takes into account the network robustness at both control and data plane layers. For given maximum values of switch-controller and controller-controller delays at the regular state (i.e., when the network is fully operational), the aim is to maximize the network robustness against a set of failure states, each state defined as a possible malicious attack to multiple network nodes. We assume that the attacker knows the data plane topology and, therefore, can adopt either one of three commonly considered node centrality attacks (based on the node degree, closeness or betweenness centralities), or an attack to the nodes which are the optimal solution of the critical node detection (CND) problem. We propose a set of robustness metrics which are used to obtain the optimal solutions for the robust CPP variant. We present a set of computational results comparing the average delays and robustness values of the robust CPP solutions against those minimizing only the average switch-controller and controller-controller delays. Moreover, the impact of using the CND based attack in the robustness evaluation of CPP solutions is also assessed in the computational results. Dorabella Santos 330485 Instituto de Telecomunicações - Pólo de Aveiro Portugal 1 Amaro F. de Sousa 3622 Institute of Telecommunications, University of Aveiro Portugal 0 Carmen Mas Machuca 168744 Technical University of Munich Germany 0 09:30 10:00 1570467857 TS3.2 SDX-based security collaboration: Extending the security reach beyond network domains 14th International Conference on Network and Service Management 2018 - Full papers Traditionally, internal networks rely on border firewalls as the first line of defence against untrusted external networks. However, customized security measures can begin further away. We have already noticed proliferating deployment of Software Defined Networking (SDN) in internal networks, and even further within the Internet core, at Internet Exchange Points (IXP). Software-Defined Internet Exchange Points (SDX) enable flexible and programmable control over the delivery of wide area network traffic. Therefore, SDX's are an appealing place to introduce security actions that span beyond the edge of internal networks. By extending security actions to SDX's, dedicated border security appliances would no longer be as overwhelmed and wide area network links would no longer deliver traffic only to be dropped at the edge of the destination networks. In this paper we present a hierarchical, logically centralized architecture enabling SDX security policies to be expressed by the Autonomous Systems (AS) as intents. Through SDX collaboration, these security intents can be compiled and installed at the closest available SDX relative to the offending source. Moreover, parallel intent compilation over multiple selected SDX's can be simultaneously executed, thus enabling a distributed security response activated at the Internet core. This approach that relies on SDX's allows faster adoption in contrast to changing all routers, or the Internet architecture. This proposed security collaboration could be used to address massive internet blackouts caused by DDoS attacks with the capacity to match the distributed force of the attacks today and future ones of even greater scale. Kristina Dzeparoska 1511216 University of Toronto Canada 0 Alberto Leon-Garcia 393083 University of Toronto Canada 0 Hadi Bannazadeh 869349 University of Toronto Canada 0 10:00 10:30 1570474828 TS3.3 All Eyes on You: Distributed Multi-Dimensional IoT Microservice Anomaly Detection 14th International Conference on Network and Service Management 2018 - Full papers The Internet of Things (IoT) is a Distributed System of cooperating Microservices (μSs). IoT services manage devices that monitor and control their environments. The interaction of the IoT with the physical environment creates strong security, privacy, and safety implications. It makes providing adequate security for IoT μSs essential. However, the complexity of IoT services makes detecting anomalous behavior difficult. We present a machine-learning based approach for modeling IoT service behavior by only observing inter-service communication. Our algorithm continuously learns μS models on distributed IoT nodes within an IoT site. Combining the learned models within and in-between IoT sites converges our μS models within short time. Sharing the resulting stable models among compute nodes enables good anomaly detection. As one application, firewalling IoT μSs becomes possible. Combining our autonomous μS modeling with firewalling enables retrofitting security to existing IoT installations. We enable retrofitting access control to existing non-secure IoT installations. Our proposed approach is resource efficient matching the requirements of the IoT. To evaluate the quality of our proposed algorithm, we show the behavior of our proposed algorithm for a set of common IoT attacks. We evaluate how domain knowledge enables us to decorrelate events on a node, and how adding context features improves the detection rate. Marc-Oliver Pahl 315764 Technical University of Munich Germany 1 Francois-Xavier Aubet 1604589 Technical University of Munich Germany 0 Keynote Session 3 Keynote Session 3 Quantifying your network: building a network Fitbit Inder Monga (Lawrence Berkeley National Lab, USA) Applications want more instrumentation from the network, and so do network engineers. As the era of ‘gut-feel networking' passes to ‘analytics-driven networking', more and more data about networks, including the constituent flows, is being tracked and retrieved. With networks becoming a effective sensor, new methods are being proposed to manage the streaming telemetry. This talk will share challenges and opportunities from a research and education network view, as we evolve the network to an analytics-driven, autonomous system. 11:00-11:45 2018-11-07T11:00:00+01:00 2018-11-07T11:45:00+01:00 Sala Convegni Keynote Session 4 Keynote Session 4 Analysis and Optimization of Networks for Flexibility Wolfgang Kellerer (Technical University of Munich, Germany) The emerging trend to softwarize networks based on concepts such as Network Virtualization, Software Defined Networking (SDN) and Network Functions Virtualization promises to increase flexibility in networking to be prepared for future demands. So far, a common understanding of flexibility is missing and flexibility is used rather as a qualitative argument for a network design choice. We present an approach towards evaluating network flexibility through a definition of a flexibility measure, which provides a quantitative analysis and a comparison of different network designs. We illustrate our approach with use case studies for dynamic function placement, resilience and flexible mobile networks. We also show how networks can be optimized for flexibility. To decrease adaptation time in flexible networks, we present results to speed up the execution of algorithms based on machine learning. 11:45-12:30 2018-11-07T11:45:00+01:00 2018-11-07T12:30:00+01:00 Sala Convegni TS4 Technical Session 4 Software Defined Networking and Network Function Virtualization 13:30-15:30 2018-11-07T13:30:00+01:00 2018-11-07T15:00:00+01:00 Sala Convegni Mohamed Faten Zhani (ETS Montreal) 13:30 13:52 1570466320 TS4.1 On the practical detection of the top-k flows 14th International Conference on Network and Service Management 2018 - Full papers Monitoring network traffic is an important building block for various management and security systems. In typical settings, the number of active flows in a network node is much larger than the number of available monitoring resources and there is no practical way to maintain per-flow state at the node. This situation gave rise to the recent interest in streaming algorithms where complex data structures are used to perform monitoring tasks like identifying the top-k flows using a constant amount of memory. However, these solutions require complicated per-packet operations, which are not feasible in current hardware or software network nodes. In this paper, we take a different approach to this problem and study the ability to perform monitoring tasks using efficient built- in counters available in current network devices. We show that by applying non-trivial control algorithms that change the filter assignments of these built-in counters at a fixed time interval, regardless of packet arrival rate, we can get accurate monitoring information. We provide an analytical study of the top-k flows problem and show, using extensive emulation over recent real traffic, that our algorithm can perform at least as well as the best-known streaming algorithms without using complex data structure or performing expensive per-packet operations. Jalil Moraney 1330141 Technion - Israel Institute of Technology Israel 1 Danny Raz 1382945 Nokia and Technion Israel 2 13:52 14:15 1570467446 TS4.2 P4NFV: An NFV Architecture with Flexible Data Plane Reconfiguration 14th International Conference on Network and Service Management 2018 - Full papers The advent of Network Function Virtualization (NFV) enhances the capability of communication networks to accommodate the dynamics that resides in the traffic demands and service requirements. In the meantime, the rise of programmable hardware equipment leads to flexible network services with performance guarantees. However, the task of NFV management is not trivial, considering the different realizations and the frequent reconfigurations of network services. P4, as a target independent data plane programming language, promises new opportunities to handle this task. We propose P4NFV, an NFV architecture that abstracts the infrastructure as a group of P4-enabled nodes, drives the nodes with P4 programs, and enables runtime data plane reconfiguration. Two approaches are proposed to reconfigure the functionality of the nodes. As a proof-of-concept, we implement a prototype based on the scenario of dynamic function deployment on the network edge. We evaluate the prototype in terms of packet forwarding latency, cpu utilization, and as a highlight, the behavior of network function migration. %migration procedure of both stateless and stateful network functions with the two reconfiguration approaches. Compared with a legacy NFV architecture that implements network functions as virtual machines, P4NFV can migrate network functions without service disruption, and only incur an acceptable performance degradation. Mu He 1394150 Technische Universität München Germany 1 Arsany Basta 823557 Technische Universität München Germany 0 Andreas Blenk 931565 Technische Universität München Germany 0 Nemanja Deric 1582630 Technische Universität München Germany 0 Wolfgang Kellerer 95259 Technische Universität München Germany 0 14:15 14:37 1570474655 TS4.3 A Northbound Interface for Software-based Networks 14th International Conference on Network and Service Management 2018 - Full papers The current shift from traditional network architectures to software-based solutions is offering new opportunities to allow network functionality to be managed in a flexible way. Substantial efforts have been invested in the recent years in the development of new network management approaches taking advantage of emerging paradigms such as software-defined networking and network function virtualization. Until now however there has not been much progress in the development of a northbound interface (NBI) linking high-level requirements (HLRs) capturing business objectives to management operations. This is a crucial functionality to facilitate faster service deployment and realization of business objectives. In this paper we extend the efforts towards the development of a NBI and propose a novel approach for the automatic decomposition of HLRs to network management operations. We demonstrate its functionality based on representative use cases and evaluate its feasibility through prototype implementation. The results obtained show that our solution can translate new technical requirements to network configurations in the order of a few seconds, as such enabling management of network functionality and services in short timescales. Daphne Tuncer 854477 Imperial College London United Kingdom (Great Britain) 1 Marinos Charalambides 830511 University College London United Kingdom (Great Britain) 0 Gioacchino Tangari 1411700 University College London United Kingdom (Great Britain) 0 George Pavlou 816589 University College London United Kingdom (Great Britain) 0 14:37 15:00 1570474756 TS4.4 On improving Service Chains Survivability Through Efficient Backup Provisioning 14th International Conference on Network and Service Management 2018 - Full papers With the growing adoption of Software Defined Networking (SDN) and Network Function Virtualization (NFV), large-scale NFV infrastructure deployments are gaining momentum. Such infrastructures are home to thousands of network Service Function Chains (SFCs), each composed of a chain of virtual network functions (VNFs) that are processing incoming traffic flows. Unfortunately, in such environments, the failure of a single node may break down several VNFs and thereby breaking many service chains at the same time. In this paper, we address this particular problem and investigate possible solutions to ensure the survivability of the affected service chains by provisioning backup VNFs that can take over in case of failure. Specifically, we propose a survivability management framework to efficiently manage SFCs and the backup VNFs. We formulate the SFC survivability problem as an integer linear program that determines the minimum number of required backups to protect all the SFCs in the system and identifies their optimal placement in the infrastructure. We also propose two heuristic algorithms to cope with the large-scale instances of the problem. Through extensive simulations of different deployment scenarios, we show that these algorithms provide near-optimal solutions with minimal computation time. Saifeddine Aidi 1604507 École de Technologie Supérieure (ÉTS Montreal) Canada 1 Mohamed Faten Zhani 148963 École de Technologie Supérieure (ÉTS Montreal), University of Quebec Canada 0 Yehia Elkhatib 351087 Lancaster University United Kingdom (Great Britain) 0 PS2 Poster Session 2 15:30-17:00 2018-11-07T15:30:00+01:00 2018-11-07T17:00:00+01:00 Sala Convegni 16:00 17:30 1570470859 PS2.1 Load Balancers for VNF chains Horizontal Scaling 14th International Conference on Network and Service Management 2018 - Full papers We present an architectural design and a reference implementation for horizontal scaling of virtual network function chains. Our solution does not require any changes to network functions and is able to handle stateful network functions for which states may depend on both directions of the traffic. We use connection-aware traffic load balancers based on hashing function to maintain mappings between connections and the dynamically changing network function chains. Our references implementation uses OpenFlow switches to route traffic to the assigned network function instances according to the load balancer decisions. We conducted extensive simulations to test the feasibility of the architecture and evaluate the performance of our implementation. Jorge Lobo 225715 ICREA & Universitat Pompeu Fabra Spain 0 Jiefei Ma 385745 Imperial College London United Kingdom (Great Britain) 0 Windhya Rankothge 1293187 Sri Lanka Institute of Information Technology Sri Lanka 0 Franck Le 168734 IBM T. J. Watson USA 0 Christian Makaya 182341 IBM Research USA 0 Mariceli Morales 1600498 Universitat Pompeu Fabra Spain 0 15:30 15:30 1570470862 PS2.2 SDN Hypervisors: How Much Does Topology Abstraction Matter? 14th International Conference on Network and Service Management 2018 - Full papers SDN network hypervisors realize the virtualization of software-defined networks. They intercept the control path between tenant controllers and their respective virtual Software-Defined Networks (SDN). Over-utilizing SDN hypervisor resources (i.e., CPU) can degrade the control plane performance of the tenants. Although many hypervisor proposals exists, a detailed performance modeling of SDN hypervisors is missing in literature. A precise modeling of the required SDN hypervisor resources, however, is crucial for predictable and reliable operation of virtual software-defined networks. In this paper, we measure and evaluate how topology abstraction can affect the SDN hypervisor CPU utilization. We consider two topology abstraction cases: the (1) transparent and (2) big-switch abstraction. Our measurements taken from a real testbed indicate that the big-switch abstraction can reduce the SDN hypervisor CPU utilization up to ~4x. Further, we evaluate different functions to model the SDN hypervisor CPU utilization based on our measurement results. Our evaluations show that a polynomial function provides the lowest fitting error. Motivated by our measurements, we conduct a first-step investigation of the impacts of topology abstraction on the Virtual Network Embedding (VNE) problem. Our initial simulation-based evaluations indicate that different topology abstraction procedures impact the results of the VNE problem. Nemanja Deric 1582630 Technische Universität München Germany 1 Amir Varasteh 1537973 Technische Universität München Germany 0 Arsany Basta 823557 Technische Universität München Germany 0 Andreas Blenk 931565 Technische Universität München Germany 0 Wolfgang Kellerer 95259 Technische Universität München Germany 0 15:30 15:30 1570473938 PS2.3 An Experimental Testbed for Managing BAN Services at the Network Edge 14th International Conference on Network and Service Management 2018 - Full papers In this article we investigate how to support Intra- or On-Body Area Network (BAN) applications with strict delay requirements using a network edge architecture. By using an SDN/NFV approach integrated with a mobile system it is possible to transmit a multimedia stream generated inside the human body to a mobile device, which then relays the information to the cloud for further processing. Therefore, we propose an edge approach to reduce overheads, thus making the support of time-constrained medical applications feasible. The proposed architecture has been implemented in a real testbed and the performance of the system were assessed to prove its feasibility. Laura Galluccio 856611 University of Catania Italy 0 Christian Grasso 1545624 University of Catania Italy 0 Sebastiano Milardo 1145287 MIT & CNIT USA 0 Giovanni Schembra 1710 University of Catania Italy 0 Elisabetta Sciacca 1138383 University of Catania Italy 0 15:30 15:30 1570474077 PS2.4 SMARTHO: A Network Initiated Handover in NG-RAN using P4-based Switches 14th International Conference on Network and Service Management 2018 - Full papers This paper deals with the design of protocols for 5G-and-beyond wireless networks. In particular, it considers a Next Generation RAN (NG-RAN), where the Base Band Unit (BBU) functions are split across a Central Unit (CU) and a Distributed Unit (DU). This paper proposes the use of Programming Protocol independent Packet Parsers (P4)-based switches between the CU and DU for processing packets exchanged between the two. We demonstrate the smart handover (SMARTHO) scheme for a mobile User Equipment (UE) that traverses a known fixed path. The idea is to perform the resource allocation in subsequent macro-cells in advance of the user's movement, by having the P4 switch spoof the behaviour of the UE. Based on an implementation using Mininet and P4BM software switches, it is seen that the proposed method leads to around 18% and 25% reduction in handover time, for two- and three-handover sequences, respectively. Phanindra Palagummi 1597257 Indian Institute of Technology Madras India 0 Krishna M. Sivalingam 5450 Indian Institute of Technology Madras India 0 15:30 15:30 1570474376 PS2.5 Beauty is in the Eye of the Smartphone Holder - A Data Driven Analysis of YouTube Mobile QoE 14th International Conference on Network and Service Management 2018 - Full papers Measuring the Quality of Experience (QoE) undergone by cellular network users has become paramount for cellular ISPs. Given its overwhelming dominance and ever-growing popularity, this paper focuses on the analysis of QoE for YouTube in mobile networks. Using a large-scale dataset of crowdsourced YouTube QoE measurements collected in smartphones with YoMoApp, we analyze the evolution of multiple relevant QoE-related metrics over time for YouTube mobile users. The dataset includes measurements from more than 360 users worldwide, spanning over the last five years. Our data-driven analysis shows a systematic performance and QoE improvement of YouTube in mobile devices over time, accompanied by an improvement of cellular network performance and by an optimization of the YouTube streaming behavior for smartphones. Nikolas Wehner 1377215 AIT Austrian Institute of Technology GmbH Austria 0 Sarah Wassermann 1604116 Inria Paris France 0 Pedro Casas 244435 Austrian Institute of Technology (AIT) Austria 0 Michael Seufert 857079 AIT Austrian Institute of Technology GmbH Austria 1 Florian Wamser 720551 University of Wuerzburg Germany 0 15:30 15:30 1570466546 PS2.6 Impact of Device Diversity on Crowdsourced Mobile Coverage Maps 14th International Conference on Network and Service Management 2018 - Short Papers Mobile coverage maps increasingly rely on user-side measurements such as those collected from crowdsourced mobile apps. These measurements inherently span a multitude of devices, differing in models and vendors, with different radio signal reception characteristics. We show measurement based evidence on the significant deviations in received signal strength distribution seen by different devices, all other factors being equal. More crucially, we examine the accuracy of coarse-grained/fine-grained measurement based mobile coverage maps as seen from a device's perspective. Our key finding is that mobile coverage maps based on measurements from a diversity of devices are still fairly reliable from a device's perspective so long as it is among the set of devices used to collect measurements. Our study also offers guidelines on ways towards reliable measurement based mobile coverage maps in presence of device diversity. Mah-Rukh Fida 1198235 The University of Edinburgh United Kingdom (Great Britain) 0 Mahesh K Marina 128084 The University of Edinburgh United Kingdom (Great Britain) 0 15:30 15:30 1570466932 PS2.7 A Method of Transport Abstraction for 5G Radio Access Networks 14th International Conference on Network and Service Management 2018 - Short Papers In the fifth generation mobile technology era, networks which cover a variety of requirements will be constructed as virtual networks (slices) by giving them appropriate resources taking into account each service's requirements. To create a slice, the management and control of a range of physical resources within a heterogeneous network is a crucial issue. One key technique to enable this is a resource abstraction method. Also, it is important to adjust the resource abstraction level to meet the service requirement parameters set by the slice orchestrators or network operators. In this paper, we have proposed abstracted resources representing services' requirement parameters in a 5G Radio Access Network (RAN), and a method for abstracting resources from many types of physical network resource. In addition, a resource allocation procedure using the proposed abstracted resources has been introduced. Akiko Nagasawa 1401235 Mitsubishi Electric Corporation Japan 1 Yukio Hirano 1596193 Mitsubishi Electric Corporation Japan 0 Kenichi Nakura 484809 Mitsubishi Electric Corporation Japan 0 Takeshi Suehiro 879657 Mitsubishi Electric Corporation Japan 2 Hiromu Sato 1596201 Mitsubishi Electric Corporation Japan 0 Seiji Kozaki 332277 Mitsubishi Electric Corp Japan 0 Kazuyuki Ishida 482485 Mitsubishi Electric Corporation Japan 0 15:30 15:30 1570467512 PS2.8 Elastic Services for Edge Computing 14th International Conference on Network and Service Management 2018 - Short Papers Edge computing enables new, low-latency services close to data producers and consumers. However, edge service management is challenged by high hardware heterogeneity and missing elasticity capabilities. To address these challenges, this paper introduces the concept of elastic services. Elastic services are situation aware and can adapt themselves to the current execution environment dynamically to adhere to their Service Level Objectives (SLOs). This adaptation is achieved through Diversifiable Programming (DivProg), a new programming model which uses function annotations as interface between the service logic, its SLOs, and the execution framework. DivProg enables developers to characterize their services in a way that allows a third-party execution framework to run them with the flow and the parametrization that conforms to changing SLOs. We develop a prototype and perform an experimental evaluation which shows that elastic services can seamlessly adapt to heterogeneous platforms and scale with a wide range of input sizes, while adhering to their SLOs with little programming effort. Jonathan Fürst 1352927 IT University of Copenhagen & NEC Laboratories Europe Denmark 1 Mauricio L. Fadel Argerich 1597197 NEC Laboratories Europe & Sapienza - Universita di Roma Germany 0 Bin Cheng 910609 NEC Laboratories Europe Germany 0 Apostolos Papageorgiou 544981 Fundació i2CAT, Internet i Innovació Digital a Catalunya Spain 0 15:30 15:30 1570470557 PS2.9 An Adaptive Policy Approach to Video Quality Assurance 14th International Conference on Network and Service Management 2018 - Short Papers Video in all its forms is probably the most important service carried on networks today and few would argue that video quality assurance is one of the most daunting network management challenges. Quite often, video optimization strategies and their decisions are an integral part of either the video protocol (e.g. dynamic adaptation of rate and quality) or the distribution systems (e.g. multi-level caching architectures). A unified method of assuring video services is a formidable task, especially as the world prepares for the adoption of 5G network concepts and the associated complexity. In parallel, policy has been proposed as an approach for managing domains in a flexible and adaptive manner. In this work, we describe our approach to use adaptive policy to externalize the goals and decision making of optimization strategies in the form of a network resource evaluation and path selection experiment aimed at video service quality assurance. In this paper, we present our approach, outline our initial implementation and discuss our preliminary results. Joseph McNamara 1600033 Athlone Institute of Technology & Ericsson Ireland 1 Sven van der Meer 1504608 Ericsson Spain 0 Liam Fallon 150427 Ericsson Ireland 0 John Keeney 174143 Ericsson Ireland 0 Enda Fallon 192539 Athlone Institute of Technology Ireland 0 15:30 15:30 1570470716 PS2.10 Virtual Network Embedding with Formal Reachability Assurance 14th International Conference on Network and Service Management 2018 - Short Papers Networks are becoming increasingly software-defined and automated. In this context, SDN and NFV allow service providers to use the network infrastructure more efficiently with reduced cost and to develop secure services. This procedure of efficient mapping of virtual networks on the substrate network is delegated to an orchestrator component of NFV that automatically manages its constituent virtualized network functions (VNFs). However, incomplete or inconsistent configuration of VNFs and service graphs may be vulnerable to potential security threats and could cause breakdown of services and of the supporting infrastructure. The main purpose of this paper is to provide an approach for allocation and formal verification that can ensure at the same time that policies such as reachability or isolation are never violated and that optimization is achieved. This ability to orchestrate and automate service validation makes assurance of reliable service delivery possible and simplifies security management tasks for network administrators. Jalolliddin Yusupov 1600317 Polytechnic University of Turin Italy 0 Guido Marchetto 123148 Politecnico di Torino Italy 0 Riccardo Sisto 145590 Politecnico di Torino Italy 0 Adlen Ksentini 100856 Eurecom France 0 15:30 15:30 1570470844 PS2.11 Towards Application of Cuckoo Filters in Network Security Monitoring 14th International Conference on Network and Service Management 2018 - Short Papers In this paper we study the feasibility of applying the recently proposed cuckoo filters to improve space efficiency for set membership testing in Network Security Monitoring, focusing on the example of threat intelligence matching. We present conceptual insights for the practical application of cuckoo filters and provide a cuckoo filter implementation allowing runtime configuration. To evaluate the practical applicability of cuckoo filters, we integrate our implementation into the Bro Network Security Monitor, compare it to traditional data structures and conduct a brief operational evaluation. We find that cuckoo filters allow remarkable memory savings, while potential performance trade-offs, caused by introducing false positives, have to be carefully evaluated on a case-by-case basis. Jan Grashöfer 1596374 Karlsruhe Institute of Technology (KIT) Germany 0 Florian Jacob 1600482 Karlsruhe Institute of Technology (KIT) Germany 0 Hannes Hartenstein 102042 Karlsruhe Institute of Technology Germany 0 TS5 Technical Session 5 Wireless and mobile networks Thursday, 8 November, 2018 09:00-10:30 2018-11-08T09:00:00+01:00 2018-11-08T10:30:00+01:00 Sala Convegni Daphne Tuncer (UCL) 09:00 09:30 1570470730 TS5.1 Trade-offs in Cache-enabled Mobile Networks 14th International Conference on Network and Service Management 2018 - Full papers Mobile data traffic demand has been growing at an unprecedented rate in the last few years. Cache-enabled mobile edge computing is known to be one of the most promising techniques to accommodate the traffic demand and alleviate the congestion at the backhaul links. However, due to limited cache capacity at the eNBs, at some parts of the network, congestion of the backhaul links and the radio resources is still possible. Thus, efficient approaches are needed in order to cache content at eNBs as well as to leverage the utilization of the resources in the mobile network while trying to avoid their congestion. In this paper, we study the trade-offs between the radio resource utilization and the backhaul link utilization in cache-enabled mobile networks. Initially, we show the trade-offs by formulating a mobility-aware joint content caching, user association, and resource allocation problem as an Integer Linear Programming problem and proposing a greedy heuristic to solve the large instances of the problem. We then propose an approach to compute radio resource and backhaul link costs, and by using the costs, we formulate a joint user association and resource allocation problem aiming at preventing network congestion, assuming that the cached content is given. The results reveal that around 10% more users get an association to the network by using the proposed algorithm. Davit Harutyunyan 1413171 FBK CREATE-NET Italy 0 Abbas Bradai 526222 XLIM Institute, University of Poitiers France 0 Roberto Riggio 146926 FBK CREATE-NET Italy 0 09:30 10:00 1570470773 TS5.2 Available Bandwidth vs. Achievable Throughput Measurements in 4G Mobile Networks 14th International Conference on Network and Service Management 2018 - Full papers Mobile broadband (MBB) networks are being increasingly used worldwide, and they are planned to steadily evolve to support new services, bigger user base, and booming machine-to-machine communications. In this scenario, performance measurements on deployed networks becomes crucial. In particular, being aware of the achievable throughput has multiple important uses ranging from path and server selection, to root-cause analysis. Throughput estimation suffers from high intrusiveness and dependence on transport and application protocol, while Available Bandwidth is a network-layer metric characterizing the spare capacity of path, not suffering from any of these shortcomings. However, ABw estimation tools have been mainly developed focusing on wired networks, with limited attention to mobile and wireless scenarios, multi-homed mobile nodes, and their peculiarities. In this work, we analyze ABw estimates and TCP achievable throughput measurements on a real 3G/4G testbed (the MONROE platform) performing tests with different providers from countries across Europe. The two metrics are compared for results and in terms of measurement intrusiveness and time cost, highlighting the non-trivial relationship between them. In particular, results show that in suitable conditions ABw estimates can be used as a proxy for TCP achievable throughput measurements, while generating much lower traffic volumes (a critical asset on MBB networks), and that policies enforced by service providers may significantly alter the difference between the two estimated metrics. Future research on how to exploit these findings is discussed as well. We published as open data packet traces and logs of the measurements. Giuseppe Aceto 388621 University of Napoli Federico II Italy 0 Fabio Palumbo 1529199 Università di Napoli Federico II Italy 1 Valerio Persico 972225 University of Napoli Federico II Italy 0 Antonio Pescapé 95375 University of Napoli Federico II Italy 0 10:00 10:30 1570470865 TS5.3 Dynamic Network Slicing for LoRaWAN 14th International Conference on Network and Service Management 2018 - Full papers One of the most important novelty in 5G is network slicing proposed as a collection of logical network functions for services running on a common physical device. In an Internet of Things (IoT) context, network resources need to be efficiently reserved and assigned for IoT devices in an isolated manner to handle and support specific Quality of Service (QoS) requirements for each slice. The focus of this paper is to investigate network slicing in LoRa network and propose a dynamic inter-slicing algorithm based on a maximum likelihood estimation that avoids resource starvation and prioritizes a slice over another depending on its QoS requirements. Moreover, we place the emphasis on a novel intra-slicing strategy that maximizes resource allocation efficiency of LoRa slices with regard to their delay requirements. After integrating an energy module for LoRa in NS3, simulation results performed in realistic LoRa scenarios highlight the utility of our dynamic network slicing proposition in providing isolation between slices with specific QoS guarantees. Samir Dawaliby 1421416 Xlim Institute, University of Poitiers France 0 Abbas Bradai 526222 XLIM Institute, University of Poitiers France 0 Yannis Pousset 1042345 Université de Poitiers France 0 Roberto Riggio 146926 FBK CREATE-NET Italy 0 Keynote Session 5 Keynote Session 5 Do Androids Dream of True Automation? Myths and Reality of Deploying Software-Defined Networking Architectures Christian Jacquenet (Orange, France) Network operators provide an ever-growing service portfolio. The diversity and the complexity of these services have been raising technical challenges for many years, not only during the service design phase but also during the service operation phase. The emergence of Software-Defined Networking (SDN) techniques such as dynamic resource allocation schemes, as well as network function virtualization techniques has often been the opportunity to make debatable promises about their so-called flexibility or their intrinsic ability to facilitate the automation of the service delivery procedures. Reality is much different. Process automation is often restricted to dynamic configuration tasks, whose steering relies upon decision-making procedures that remain "manually declarative": the data that are used to feed the computation logic that will drive the execution of configuration tasks are statically declared. In addition, this rather embryonic automation only deals with tasks that remain local to a device in the detriment of a global, systemic view that would be able to guarantee the global consistency of the actions initiated to deliver a service. 11:00-11:45 2018-11-08T11:00:00+01:00 2018-11-08T11:45:00+01:00 Sala Convegni Keynote Session 6 Keynote Session 6 5G+ - the Journey of ICDT Deep Convergence Chih-Lin I (China Mobile Research Institute, China) As highly anticipated 5G phase one deployment is speeding up on a global scale, successors of 5G and their key features are being sought. This keynote will highlight China Mobile's 5G R&D themes (Green & Soft) that began with multiple initiatives to "Rethink the Fundamentals". It led to an E2E Soft architecture reflected in the 5G NR specifications. Subsequent pursuit of "Open & Smart" themes of future wireless communication systems will then be elaborated. ‘Open' embraces open interface, white-box hardware, and open source software, while ‘Smart' refers to embedded intelligence in RAN with wireless big data and machine learning. It has become increasingly essential for us to rethink our Ecosystem, SDO Operation, System Models and Algorithms. Preliminary exploration of artificial intelligence and its associated reference architecture will be introduced. The progression from Green & Soft to Open & Smart is expected to bring forth the necessary transformation of our ecosystem amidst a true and deep ICDT convergence. 11:45-12:30 2018-11-08T11:45:00+01:00 2018-11-08T12:30:00+01:00 Sala Convegni TS6 Technical Session 6 Resource Allocation 13:30-15:00 2018-11-08T13:30:00+01:00 2018-11-08T15:00:00+01:00 Sala Convegni Danny Raz (Nokia and Technion) 13:30 14:00 1570467863 TS6.1 Congestion-Constrained Virtual Link Embedding with Uncertain Demands 14th International Conference on Network and Service Management 2018 - Full papers Network virtualization enables multiple virtual networks to co-exist on the same physical network. Each virtual network requires specific amounts of physical network resources such as node processing and link bandwidth. The problem of mapping virtual resource requirements to physical resources is extensively studied in the literature under the assumption that resource demands of virtual networks are known deterministically. In real deployments though, resource demands include significant uncertainty and fluctuate over time. This paper considers the problem of mapping virtual links to physical network paths subject to a constraint on each virtual link congestion probability under the assumption that bandwidth demands of virtual links are uncertain. A general uncertainty model is considered, where bandwidth demands are described by random variables for which only the mean and variance (or a range) are known. We formulate the problem as a nonlinear optimization problem, which is shown to be non-convex. Consequently, we develop an approximate formulation that results in a second-order cone program (SOCP) that can be solved efficiently even for large networks. We then provide simulation as well as Mininet experimental results to show the utility and efficiency of our exact and approximate models in various network scenarios. We apply our models to commonly studied USA and EON networks as well as randomly generated large networks. Our results show that both models are able to satisfy the link congestion constraint, and that the approximate model is very close to the exact model. Fatemeh Hosseini 1582325 University of Calgary Canada 1 Alexander James 1593794 University of Calgary Canada 0 Majid Ghaderi 253929 University of Calgary Canada 0 14:00 14:30 1570470361 TS6.2 UNiS: A User-space Non-intrusive Workflow-aware Virtual Network Function Scheduler 14th International Conference on Network and Service Management 2018 - Full papers Network Function Virtualization (NFV) has gained a significant research interest in both academia and industry since its inception in the late 2012. One of the key research issues in NFV is the development of systems for building Virtual Network Functions (VNFs) capable of meeting the performance requirements of enterprise and telecommunication networks. New packet processing models leveraging kernel bypass I/O and poll-mode processing have gained popularity for building high performance VNFs because of their simple programming model and very low I/O overhead. However, a major drawback of such poll-mode processing is the inefficient use of CPU resources. Existing CPU schedulers are ill-suited for VNFs due to their inability to capture the actual processing cost of a poll-mode VNF, hence, cannot rightsize CPU allocation. This is further exacerbated by their inability to consider VNF processing order when VNFs are chained to form Service Function Chains (SFCs). The state-of-the-art solutions proposed for VNF scheduling are intrusive, i.e., requiring the VNFs to be built with scheduler specific libraries or having carefully selected scheduling checkpoints. This highly restricts the VNFs that can properly work with such schedulers. In this paper, we present UNiS: a User-space Non-intrusive work-flow aware VNF Scheduler. Unlike existing approaches, UNiS does not require VNF modifications and treats the poll-mode VNFs as a black box, hence, is non-intrusive. UNiS is also workflow-aware, i.e., maintains SFC processing order while scheduling VNFs. Testbed experiments show that UNiS is able to achieve a throughput within 90% (for synthetic traffic load) and 98% (for real data center traffic trace) of the achievable throughput using an intrusive co-operative scheduler. Anthony Anthony 1600508 University of Waterloo Canada 1 Shihabur Rahman Chowdhury 341839 University of Waterloo Canada 0 Tim Bai 1600509 University of Waterloo Canada 0 Raouf Boutaba 5035 University of Waterloo Canada 0 Jérôme François 235420 INRIA Nancy Grand Est France 0 14:30 15:00 1570474734 TS6.3 Adaptive Flag-Based Signaling for Distributed Spectrum Assignment in Elastic Optical Networks 14th International Conference on Network and Service Management 2018 - Full papers The emergence of flexible frequency grids and distance adaptive modulation transmission methodologies in Elastic Optical Networks (EONs) necessitates novel and efficient Routing and Spectrum Assignment (RSA) and signaling mechanisms to support them. In this paper, we propose an adaptive flag-based signaling mechanism for distributed spectrum allocation to maximize lightpath connection establishment by lowering blocking probability (BP). Our numerical results indicate significant improvement in the rate of success in establishing dynamically arriving connections. We also proposed an enhanced adaptive approach based on hop count with data-driven learning to further reduce the blocking probability of connections Amer Al Baidhani 1439567 University of Cincinnati USA 0 Gokhan Sahin 136794 Miami University USA 0 Donald Ucci 1439650 Miami University USA 0 DEP Distinguished Experts Panel The Advent of Network Automation at last? Prosper Chemouil (Orange, France), Alex Clemm (Huawei, USA), Christian Jacquenet (Orange, France), Wolfgang Kellerer (Technical University of Munich, Germany) From the early days of networking, automation has been a goal to efficiently and safely run networks. Automatic control congestion mechanisms had been implemented to protect network equipment from overload. Dynamic routing and flow control mechanisms were proposed to maximize throughput overall. Expert systems had been implemented too in the early 90's but wide deployment did not arise. Such automated approaches were mostly suggested for network management and improved to address anomaly detection. Over the years, network automation has been addressed with various paradigms such as programmable and active networks which were considered under the framework of autonomic networking. More recently, the use of Artificial Intelligence made the concept evolve towards to cognitive networking, while SDN has provided an architectural framework for self-driven networks. Though studies and proof-of-concepts have been widely proposed and showcased, the Big Day for network Automation is still to come. Meanwhile recent advances are showing that we may soon reach a turning point in network automation: constraints imposed by the wide use of virtualization and cloud computing techniques and progress in data science based upon the use of different types of data and their processing through machine learning may provide solutions to address network and service complexity. The panel is intended to address evolution and will cover the following issues: - How network automation has evolved since initial ideas (automation, autonomic, active, cognitive, intelligent, etc…)? - What is the impact of virtualization and programmability concepts on Network Automation? - What the current ingredients for Network Automation: policy-based, Intent-based, data-based? - How Network Automation would be deployed at the edge, in radio access networks or for IoT services? - How the emergence of Open Source Network Automation Platforms is helping or hindering deployment? - How AI is expected to help Network Automation? 15:30-17:00 2018-11-08T15:30:00+01:00 2018-11-08T17:00:00+01:00 Sala Convegni Tutorial 3 Tutorial 3 Blockchain and Smart Contracts - From Theory to Practice Bruno Rodrigues, Roman Blum, Thomas Bocek, Burkhard Stiller Friday, 9 November, 2018 09:00-17:00 2018-11-09T09:00:00+01:00 2018-11-09T17:00:00+01:00 Sala Bisogno