{"title":"Hardware-accelerator aware VNF-chain recovery","authors":"G. Sharma, W. Tavernier, D. Colle, M. Pickavet","doi":"10.1109/DRCN48652.2020.1570604216","DOIUrl":"https://doi.org/10.1109/DRCN48652.2020.1570604216","url":null,"abstract":"Hardware-accelerators in Network Function Virtualization (NFV) environments have aided telecommunications companies (telcos) to reduce their expenditures by offloading compute-intensive VNFs to hardware-accelerators. To fully utilize the benefits of hardware-accelerators, VNF-chain recovery models need to be adapted. In this paper, we present an ILP model for optimizing prioritized recovery of VNF-chains in heterogeneous NFV environments following node failures. We also propose an accelerator-aware heuristic for solving prioritized VNF-chain recovery problems of large-size in a reasonable time. Evaluation results show that the performance of heuristic matches with that of ILP in regard to restoration of high and medium priority VNF-chains and a small penalty occurs only for low-priority VNF-chains.","PeriodicalId":334421,"journal":{"name":"2020 16th International Conference on the Design of Reliable Communication Networks DRCN 2020","volume":"2014 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133975280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Arunaba Sen, Chenyang Zhou, Anisha Mazumder, Arun Das, K. Basu, K. Walkowiak
{"title":"On the Number of Steiner Trees in a Graph","authors":"Arunaba Sen, Chenyang Zhou, Anisha Mazumder, Arun Das, K. Basu, K. Walkowiak","doi":"10.1109/DRCN48652.2020.1570613458","DOIUrl":"https://doi.org/10.1109/DRCN48652.2020.1570613458","url":null,"abstract":"In a number of networking problems, one needs to find multiple paths between a source-destination node pair or multiple trees spanning all (or some of) the nodes of the network. Accordingly, a number of algorithms for generating (i) k shortest paths between a specified source-destination node pair, (ii) k Spanning trees, (iii) k Steiner trees, etc. are used in networking literature. However, any network G = (V, E) will have a specific number of Spanning Trees. Suppose that for a given network this number is p. In case if one attempts to create k Spanning Trees where k > p, the effort will end up in failure, as k Spanning trees do not exist for the network. Thus, before embarking of creating k Spanning trees, one needs to make sure that k < p. In case of Spanning trees, it is easy to verify that k < p, as Cayley’s formula provides the number of labeled Spanning Trees of a complete graph. However, if one wants to generate multiple trees spanning only a specified subset of nodes, Cayley’s formula will not be helpful, as in this case one needs to create Steiner trees by connecting the specified subset of nodes. The goal of this paper is to find a counterpart of Cayley’s formula for the Steiner trees. To the best of our knowledge, no such formula is known for the number of Steiner Trees in a complete graph with n nodes and p terminal nodes. In this paper, we first provide a formula for the number of Steiner Trees of a complete graph, when p = 2 and p = 3. For p ≥ 4 we provide two algorithms, of which the first one computes the number of Steiner Trees, and the second one generates all the Steiner Trees of a complete graph with n nodes and p terminal nodes. The complexity of the first algorithm is O(np) and the second algorithm generates a Steiner Tree every O(n) unit of delay. It may be noted that the second algorithm can be used for any graphs, not only for the complete graphs.","PeriodicalId":334421,"journal":{"name":"2020 16th International Conference on the Design of Reliable Communication Networks DRCN 2020","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114992031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sugang Xu, N. Yoshikane, M. Shiraiwa, Y. Hirota, T. Tsuritani, Sifat Ferdousi, Y. Awaji, N. Wada, B. Mukherjee
{"title":"Toward Disaster-Resilient Optical Networks with Open and Disaggregated Subsystems [Invited]","authors":"Sugang Xu, N. Yoshikane, M. Shiraiwa, Y. Hirota, T. Tsuritani, Sifat Ferdousi, Y. Awaji, N. Wada, B. Mukherjee","doi":"10.1109/DRCN48652.2020.1570628731","DOIUrl":"https://doi.org/10.1109/DRCN48652.2020.1570628731","url":null,"abstract":"Novel open and disaggregated optical-networking technologies promise to enhance multi-vendor interoperability thanks to their open interfaces in both data-plane and control/management-plane (C/M-plane). From the viewpoint of disaster resilience in optical networks, such interoperability will significantly improve the flexibility in product selection with regard to replacing damaged subsystems with products of different vendors. In this paper, we discuss various approaches for rapid post-disaster recovery in optical networks (including legacy optical networks) employing disaggregated subsystems, namely, the emergency first-aid unit (FAU) with open application programming interfaces and protocols. We address the following problems (and introduce the solutions that we are currently investigating): (1) how to take advantage of the new disaggregated resources and surviving legacy optical resources to achieve early recovery, (2) how to achieve integrated control of FAUs and non-FAU legacy ROADMs, and (3) how to quickly recreate the lost optical performance monitoring (OPM) capability with FAUs and perform a robust telemetry under the restricted bandwidth in the degraded C/M-plane networks.","PeriodicalId":334421,"journal":{"name":"2020 16th International Conference on the Design of Reliable Communication Networks DRCN 2020","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128560449","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Framework for Network Function Decomposition and Deployment","authors":"Daniele Moro, G. Verticale, A. Capone","doi":"10.1109/DRCN48652.2020.1570613823","DOIUrl":"https://doi.org/10.1109/DRCN48652.2020.1570613823","url":null,"abstract":"Network Function Virtualization (NFV) enables fast provisioning of packet processing logic on general purpose CPUs. This approach, however, does not scale well to very high speed traffic. Programmable hardware solutions, including those based on programmable switches, are emerging as an option for accelerating and scaling network functions. Unfortunately, every type of programmable hardware has specific characteristics that do not make it suitable for running all possible functions. We argue that an efficient strategy is decomposing network functions into components that can run on CPUs or that can be offloaded to specific programmable hardware depending on their characteristics.This paper presents a preliminary work on a framework for automating the decomposition and deployment of network functions. The framework includes an orchestrator that chooses the best decomposition according to the traffic demands, the network topology and other constraints. It also provides a tool to combine multiple functions into a single P4 program that can be deployed to a programmable switch. Finally, the framework comprises a set of tools to deploy the network functions either as containers running in a data center or as programs loaded in a programmable switch.We present numerical results to highlight the advantages of partially offloading decomposed VNFs to programmable hardware over a pure software solution. We also highlight the robustness of the approach showing how the model reacts in case of network failures.","PeriodicalId":334421,"journal":{"name":"2020 16th International Conference on the Design of Reliable Communication Networks DRCN 2020","volume":"359 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122748112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Resilience of 5G Networks in the Presence of Unlicensed Spectrum and Non-Terrestrial Networks","authors":"Rohit Abhishek, D. Tipper, D. Medhi","doi":"10.1109/DRCN48652.2020.1570604438","DOIUrl":"https://doi.org/10.1109/DRCN48652.2020.1570604438","url":null,"abstract":"The 5G wireless network is geared to provide greater capacity, increased energy efficiency and lower latency. A critical issue in service delivery is to provide resilience in 5G networks. In this work, we envision an environment where 5G network resilience is addressed in the presence of unlicensed spectrum and non-terrestrial networks. We present a framework for network survivability with network virtualization with multiple providers, and the use of unlicensed spectrum band and non-terrestrial network (NTN); this is done along with a self organizing ad hoc network among the gNBs that may use a secondary provider for network resilience when the aggregation network and the backhaul network fails. In this architecture, we present an optimization model for survivability for a 5G networks provider (primary provider) that may also use a secondary provider in the event of a failure along with unlicensed spectrum and NTN. Our simulations show (1) the trade-off between using a primary providers own network or rely on auxiliary capacity from the secondary provider, and (2) the use of unlicensed spectrum band and NTN enhances resilience of the network.","PeriodicalId":334421,"journal":{"name":"2020 16th International Conference on the Design of Reliable Communication Networks DRCN 2020","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125645885","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"High-Availability Service Chain Realization Theory","authors":"Sidharth Sharma, A. Gumaste, M. Tatipamula","doi":"10.1109/DRCN48652.2020.1570611001","DOIUrl":"https://doi.org/10.1109/DRCN48652.2020.1570611001","url":null,"abstract":"We propose a novel approach to realize service chains with specific availability and latency requirements by dimensioning a NFV infrastructure (NFVI). Our contribution leads to a theory that is based on extending SDN capacity computation to the domain of NFV. We do so by expanding upon the requirements of availability of service chains (SCs), projecting these onto a NFVI. A key contribution of our theoretical model is the inclusion of delay bounds while traversing a SC. The combination of availability, delay bounds and computing SC granularity is achieved through modeling SCs as directed acyclic graphs in a NFVI, and acted upon using network calculus, eventually leading to a SC realization theorem. This SC realization theorem answers a question as to how many SCs of a particular type and granularity can be provisioned on a particular DC. A second fast approach towards computing SCs realization on a NFVI is also presented. The theorems are verified numerically for DCs of size upto 12,228 servers (nodes). Results for scalability and latency are presented.","PeriodicalId":334421,"journal":{"name":"2020 16th International Conference on the Design of Reliable Communication Networks DRCN 2020","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131663213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"On the Resilience of the NFV-MANO: An Availability Model of a Cloud-native Architecture","authors":"Besmir Tola, Yuming Jiang, B. Helvik","doi":"10.1109/DRCN48652.2020.1570604378","DOIUrl":"https://doi.org/10.1109/DRCN48652.2020.1570604378","url":null,"abstract":"With Network Function Virtualization (NFV), the management and orchestration of network services require a new set of functionalities to be added on top of legacy models of operation. Due to the introduction of the virtualization layer and the decoupling of the network functions and their running infrastructure, the operation models need to include new elements like virtual network functions (VNFs) and a new set of relationships between them and the NFV Infrastructure (NFVI). The NFV Management and Orchestration (MANO) framework plays the key role in managing and orchestrating the NFV infrastructure, network services and the associated VNFs. Failures of the MANO hinders the network ability to react to new service requests or events related to the normal lifecycle operation of network services. Thus, it becomes extremely important to ensure a high level of availability for the MANO architecture. The goal of this work is to model, analyze, and evaluate the impact that different failure modes have on the MANO availability. A model based on Stochastic Activity Networks (SANs), derived from current standard-compliant microservice-based implementations, is proposed as a case study. The case study is used to quantitatively evaluate the steady-state availability and identify the most important parameters influencing the system availability for different deployment configurations.","PeriodicalId":334421,"journal":{"name":"2020 16th International Conference on the Design of Reliable Communication Networks DRCN 2020","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115230306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Bartlomiej Ostrowski, M. Pióro, A. Tomaszewski, M. Nowak, I. Kalesnikau
{"title":"Throughput vs. Resilience in Multi-hop Wireless Sensor Networks with Periodic Packet Traffic","authors":"Bartlomiej Ostrowski, M. Pióro, A. Tomaszewski, M. Nowak, I. Kalesnikau","doi":"10.1109/DRCN48652.2020.1570610581","DOIUrl":"https://doi.org/10.1109/DRCN48652.2020.1570610581","url":null,"abstract":"In this paper we consider multi-hop wireless sensor networks using time division multiple access to radio channel. It is assumed that packets, periodically generated at the sensors at the beginning of each consecutive time frame, are to be delivered to (Internet) gateways. Although in the nominal network condition it is sufficient when the packet streams produced by each sensor reach one of the gateways, from resilience viewpoint it would be advantageous to deliver them to multiple gateways, preferably using the broadcast capability of wireless nodes. Clearly, under the assumption of periodic packet traffic, a longer frame must be used when packets are to be delivered (without losses) to multiple gateways rather to single gateways. Since traffic throughput (one packet per stream per frame) is inversely proportional to the frame length, a clear tradeoff between resilience and traffic throughput arises. In the paper we present a numerical study, using a modification of our previous optimization model for frame length minimization in the considered kind of networks, that illustrates the relation between the frame length and the number of gateways each packet stream is supposed to reach. A general conclusion is that the frame length increase is non-negligible when multiple gateways are assumed, and depends heavily on the way the resulting multicast traffic is handled and how the particular subsets of gateways are selected for particular streams.","PeriodicalId":334421,"journal":{"name":"2020 16th International Conference on the Design of Reliable Communication Networks DRCN 2020","volume":"40 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116350802","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Unified View to Machine Learning and Control for Measurement-based Equivalent Bandwidth","authors":"M. Mongelli, M. Muselli, M. Marchese","doi":"10.1109/DRCN48652.2020.1570609737","DOIUrl":"https://doi.org/10.1109/DRCN48652.2020.1570609737","url":null,"abstract":"This paper outlines a unified view of machine learning and control for the optimization of a communication system. The problem of equivalent bandwidth is taken as a reference. A dedicated classification technique is used to derive insights into the structure of the problem by means of boolean rules over the variables of the system. The approach is of particular interest for many settings in which only measurements of the performance are available. Simulations corroborate the quality of the proposed technique.","PeriodicalId":334421,"journal":{"name":"2020 16th International Conference on the Design of Reliable Communication Networks DRCN 2020","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123083737","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Knowledge-Based Collective Self-learning for Alarm Prediction in Real Multi-Domain Autonomous Optical Networks","authors":"Xiangdong Xing, Yongli Zhao, Yajie Li, Jie Zhang","doi":"10.1109/DRCN48652.2020.1570611066","DOIUrl":"https://doi.org/10.1109/DRCN48652.2020.1570611066","url":null,"abstract":"In this paper, a collective self-learning method based on knowledge sharing is proposed to predict alarms in multi-domain autonomous optical networks. The well-considered architecture is rendered, together with various alternatives for combining machine learning (ML) knowledge. The proposed method has been tested in the commercial large-scale multidomain network with 274 nodes and 487 links. Experimental results show that it can achieve high accuracy for alarm prediction. In addition, it can achieve similar performance with much better flexibility than a collective scheme based on training data sharing as well as more superior accuracy and robustness than an individual ML model.","PeriodicalId":334421,"journal":{"name":"2020 16th International Conference on the Design of Reliable Communication Networks DRCN 2020","volume":"12 18","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132815507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}