Pub Date : 2019-10-01DOI: 10.1109/RNDM48015.2019.8949121
K. Manousakis, T. Panayiotou, P. Kolios, Ioannis Tomkos, G. Ellinas
This work considers lightpath provisioning in elastic optical networks with traffic demand variations while accounting for the impact of jamming attacks. Traffic requests are modeled based on their variation in traffic and a number of network traffic scenarios are pre-computed. Variations in traffic are used in order to design the network with improved resilience to jamming attacks in the general case, without considering a specific traffic matrix. The mathematical formulation and heuristic algorithms proposed in this work jointly consider the routing and spectrum allocation problem for the pre-computed network scenarios, aiming to minimize the required lightpath reallocations and the number of wavelength selective switches (WSSs) placed at specific network nodes/ports so as to minimize the impact of jamming attacks. Performance results demonstrate the benefits of this approach in terms of required lightpath reallocations, number of WSSs, as well as computational time required for dynamically reconfiguring the network when considering traffic demand variations.
{"title":"Attack-aware Lightpath Provisioning in Elastic Optical Networks with Traffic Demand Variations","authors":"K. Manousakis, T. Panayiotou, P. Kolios, Ioannis Tomkos, G. Ellinas","doi":"10.1109/RNDM48015.2019.8949121","DOIUrl":"https://doi.org/10.1109/RNDM48015.2019.8949121","url":null,"abstract":"This work considers lightpath provisioning in elastic optical networks with traffic demand variations while accounting for the impact of jamming attacks. Traffic requests are modeled based on their variation in traffic and a number of network traffic scenarios are pre-computed. Variations in traffic are used in order to design the network with improved resilience to jamming attacks in the general case, without considering a specific traffic matrix. The mathematical formulation and heuristic algorithms proposed in this work jointly consider the routing and spectrum allocation problem for the pre-computed network scenarios, aiming to minimize the required lightpath reallocations and the number of wavelength selective switches (WSSs) placed at specific network nodes/ports so as to minimize the impact of jamming attacks. Performance results demonstrate the benefits of this approach in terms of required lightpath reallocations, number of WSSs, as well as computational time required for dynamically reconfiguring the network when considering traffic demand variations.","PeriodicalId":120852,"journal":{"name":"2019 11th International Workshop on Resilient Networks Design and Modeling (RNDM)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128420580","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}
Pub Date : 2019-10-01DOI: 10.1109/RNDM48015.2019.8949156
H. Saeed, R. Reyes, T. Bauschert, M. Gunkel, Felix Wissel
Bandwidth variable transponders (BVTs) have the capability to adjust the line-rates of lightpaths in optical transport networks. This feature is relevant for the restoration of IP over optical networks where the capacity of the IP links is provisioned by lightpaths. When a fibre cut occurs, the affected IP links are restored by re-routing the disrupted lightpaths over usually longer protection paths. In most cases, to adapt to the physical limitations of these paths, the BVTs that implement the affected lightpaths have to reduce their line-rates. This results in IP links with reduced capacities which may cause congestion in the IP layer. Today, due to the lack of information exchange between the IP and the optical control planes, the IP layer is not aware of the capacity reduction and cannot trigger automatic traffic re-routing so as to avoid congestion. Therefore, the policy applied today is to shut-off all IP links affected by a line-rate reduction. In this paper we study alternatives to this policy by investigating re-routing strategies in the IP layer that avoid shutting-off all IP links affected. The assumption is made that in future a software defined network (SDN) control plane enables coordination between the IP and optical layers. The efficiency of the strategies is assessed by calculating their spare capacity requirements, i.e. the extra capacity required in the optical layer to carry the IP traffic in case of single link fibre cuts. In our case study we consider a network scenario from Deutsche Telekom. The results show that the rerouting strategies lead to significant capacity savings compared to the traditional shut-off approach.
{"title":"Spare Capacity Dimensioning for different Re-Routing Strategies in Transport Networks","authors":"H. Saeed, R. Reyes, T. Bauschert, M. Gunkel, Felix Wissel","doi":"10.1109/RNDM48015.2019.8949156","DOIUrl":"https://doi.org/10.1109/RNDM48015.2019.8949156","url":null,"abstract":"Bandwidth variable transponders (BVTs) have the capability to adjust the line-rates of lightpaths in optical transport networks. This feature is relevant for the restoration of IP over optical networks where the capacity of the IP links is provisioned by lightpaths. When a fibre cut occurs, the affected IP links are restored by re-routing the disrupted lightpaths over usually longer protection paths. In most cases, to adapt to the physical limitations of these paths, the BVTs that implement the affected lightpaths have to reduce their line-rates. This results in IP links with reduced capacities which may cause congestion in the IP layer. Today, due to the lack of information exchange between the IP and the optical control planes, the IP layer is not aware of the capacity reduction and cannot trigger automatic traffic re-routing so as to avoid congestion. Therefore, the policy applied today is to shut-off all IP links affected by a line-rate reduction. In this paper we study alternatives to this policy by investigating re-routing strategies in the IP layer that avoid shutting-off all IP links affected. The assumption is made that in future a software defined network (SDN) control plane enables coordination between the IP and optical layers. The efficiency of the strategies is assessed by calculating their spare capacity requirements, i.e. the extra capacity required in the optical layer to carry the IP traffic in case of single link fibre cuts. In our case study we consider a network scenario from Deutsche Telekom. The results show that the rerouting strategies lead to significant capacity savings compared to the traditional shut-off approach.","PeriodicalId":120852,"journal":{"name":"2019 11th International Workshop on Resilient Networks Design and Modeling (RNDM)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124080603","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}
Pub Date : 2019-10-01DOI: 10.1109/RNDM48015.2019.8949107
T. Kumar, An Braeken, Vidhya Ramani, Ijaz Ahmad, E. Harjula, M. Ylianttila
Internet of Things (IoT), together with fifth-generation (5G) mobile systems and related enabling communication technologies, constantly contributing for the enhancement in various industrial applications and driving towards the vision of complete industrial automation. In this context, both Blockchain and Edge computing are considered as promising technologies to fulfill the vision and requirements for the future Industry 4.0. The integration of these technologies together with massive scale Industrial IoT (IIoT) applications would be vital in addressing several key challenges related to e.g. latency, scalability and security of data processing and sharing, among others. However, such complex IIoT systems are constantly exposed to security threats from various sources. This paper focuses on identifying the major security challenges in the IIoT blockchain-edge related networks. In addition, some potential security solutions are presented in order to overcome these challenges.
{"title":"SEC-BlockEdge: Security Threats in Blockchain-Edge based Industrial IoT Networks","authors":"T. Kumar, An Braeken, Vidhya Ramani, Ijaz Ahmad, E. Harjula, M. Ylianttila","doi":"10.1109/RNDM48015.2019.8949107","DOIUrl":"https://doi.org/10.1109/RNDM48015.2019.8949107","url":null,"abstract":"Internet of Things (IoT), together with fifth-generation (5G) mobile systems and related enabling communication technologies, constantly contributing for the enhancement in various industrial applications and driving towards the vision of complete industrial automation. In this context, both Blockchain and Edge computing are considered as promising technologies to fulfill the vision and requirements for the future Industry 4.0. The integration of these technologies together with massive scale Industrial IoT (IIoT) applications would be vital in addressing several key challenges related to e.g. latency, scalability and security of data processing and sharing, among others. However, such complex IIoT systems are constantly exposed to security threats from various sources. This paper focuses on identifying the major security challenges in the IIoT blockchain-edge related networks. In addition, some potential security solutions are presented in order to overcome these challenges.","PeriodicalId":120852,"journal":{"name":"2019 11th International Workshop on Resilient Networks Design and Modeling (RNDM)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134303827","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}
Pub Date : 2019-10-01DOI: 10.1109/RNDM48015.2019.8949084
P. Romero
Stochastic Binary Systems (SBS) are mathematical models of multi-component on-off systems subject to random failures on its components. In this abstract setting, an SBS can model a network, an information system or the operation of an industrial automatic process, among many others. Given the importance of its applications, recent works from the literature aim to understand the system reliability and combinatorics of some distinguished sub-classes of SBS. In this paper we revisit the progress in this field. The most challenging problems in the analysis and practice of SBS are also discussed.
{"title":"Challenges in Stochastic Binary Systems","authors":"P. Romero","doi":"10.1109/RNDM48015.2019.8949084","DOIUrl":"https://doi.org/10.1109/RNDM48015.2019.8949084","url":null,"abstract":"Stochastic Binary Systems (SBS) are mathematical models of multi-component on-off systems subject to random failures on its components. In this abstract setting, an SBS can model a network, an information system or the operation of an industrial automatic process, among many others. Given the importance of its applications, recent works from the literature aim to understand the system reliability and combinatorics of some distinguished sub-classes of SBS. In this paper we revisit the progress in this field. The most challenging problems in the analysis and practice of SBS are also discussed.","PeriodicalId":120852,"journal":{"name":"2019 11th International Workshop on Resilient Networks Design and Modeling (RNDM)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123739154","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}
Pub Date : 2019-10-01DOI: 10.1109/RNDM48015.2019.8949117
Luís Alçada-Almeida, Teresa Gomes, C. H. Antunes
Phasor Measurement Units (PMU) are specific equipment that measure phasors, magnitude and frequency of voltage and current in key points of a power grid, allowing for a more effective management of network assets. We start by presenting some formulations for the problem of determining the location to minimize the number of PMUs installed, ensuring that all buses (nodes of the power grid) are visible by at least one PMU. The formulations take also into account the existence of Zero Injection Buses (ZIBs) and the requirement of $(mathcal{N}-1)$ redundancy to ensure that critical buses (e.g. buses with a generator) can be observed by at least two PMUs. This guarantees, in case one out of $mathcal{N}$ installed PMUs fails, all critical buses are still observable. We assume that the communications network topology follows the topology of the power grid. With the goal to protect the communication infrastructure against failures and natural disasters, we propose models that minimize the number of Shared Risk Link Groups (SRLGs) and links in the path pair of each PMU to a pair of Phasor Data Concentrators (PDCs). We then modify the formulation of the PMU location problem to improve the resilience of the communication between the PMUs and the PDCs, namely by avoiding to place PMUs in buses of degree one and also by penalizing the placement of PMUs in buses that will result in path pairs to the PDCs which are not SRLG-disjoint. Results are presented to illustrate the impact of the modified PMU location model on the resilience of the communication network.
{"title":"Optimization of PMU Location and Communications in a Power Grid","authors":"Luís Alçada-Almeida, Teresa Gomes, C. H. Antunes","doi":"10.1109/RNDM48015.2019.8949117","DOIUrl":"https://doi.org/10.1109/RNDM48015.2019.8949117","url":null,"abstract":"Phasor Measurement Units (PMU) are specific equipment that measure phasors, magnitude and frequency of voltage and current in key points of a power grid, allowing for a more effective management of network assets. We start by presenting some formulations for the problem of determining the location to minimize the number of PMUs installed, ensuring that all buses (nodes of the power grid) are visible by at least one PMU. The formulations take also into account the existence of Zero Injection Buses (ZIBs) and the requirement of $(mathcal{N}-1)$ redundancy to ensure that critical buses (e.g. buses with a generator) can be observed by at least two PMUs. This guarantees, in case one out of $mathcal{N}$ installed PMUs fails, all critical buses are still observable. We assume that the communications network topology follows the topology of the power grid. With the goal to protect the communication infrastructure against failures and natural disasters, we propose models that minimize the number of Shared Risk Link Groups (SRLGs) and links in the path pair of each PMU to a pair of Phasor Data Concentrators (PDCs). We then modify the formulation of the PMU location problem to improve the resilience of the communication between the PMUs and the PDCs, namely by avoiding to place PMUs in buses of degree one and also by penalizing the placement of PMUs in buses that will result in path pairs to the PDCs which are not SRLG-disjoint. Results are presented to illustrate the impact of the modified PMU location model on the resilience of the communication network.","PeriodicalId":120852,"journal":{"name":"2019 11th International Workshop on Resilient Networks Design and Modeling (RNDM)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129536202","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}
Pub Date : 2019-10-01DOI: 10.1109/RNDM48015.2019.8949119
Zhidong He, Peng Sun, P. Mieghem
Network recoverability refers to the ability of a network to return to a desired performance level after suffering malicious attacks or random failures. This paper proposes a general topological approach and recoverability indicators to measure the network recoverability in two scenarios: 1) recovery of damaged connections and 2) any disconnected pair of nodes can be connected to each other. Our approach presents the effect of the random attack and recovery processes on the network performance by the robustness envelopes of realizations and the histograms of two recoverability indicators. By applying the effective graph resistance and the network efficiency as robustness metrics, we employ the proposed approach to assess 10 realworld communication networks. Numerical results verify that the network recoverability is coupled to the network topology, the robustness metric and the recovery strategy. We also show that a greedy recovery strategy could provide a near-optimal recovery performance for the investigated robustness metrics.
{"title":"Topological Approach to Measure Network Recoverability","authors":"Zhidong He, Peng Sun, P. Mieghem","doi":"10.1109/RNDM48015.2019.8949119","DOIUrl":"https://doi.org/10.1109/RNDM48015.2019.8949119","url":null,"abstract":"Network recoverability refers to the ability of a network to return to a desired performance level after suffering malicious attacks or random failures. This paper proposes a general topological approach and recoverability indicators to measure the network recoverability in two scenarios: 1) recovery of damaged connections and 2) any disconnected pair of nodes can be connected to each other. Our approach presents the effect of the random attack and recovery processes on the network performance by the robustness envelopes of realizations and the histograms of two recoverability indicators. By applying the effective graph resistance and the network efficiency as robustness metrics, we employ the proposed approach to assess 10 realworld communication networks. Numerical results verify that the network recoverability is coupled to the network topology, the robustness metric and the recovery strategy. We also show that a greedy recovery strategy could provide a near-optimal recovery performance for the investigated robustness metrics.","PeriodicalId":120852,"journal":{"name":"2019 11th International Workshop on Resilient Networks Design and Modeling (RNDM)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128219532","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}
Pub Date : 2019-10-01DOI: 10.1109/RNDM48015.2019.8949088
A. Valentini, Bálazs Vass, Jorik Oostenbrink, Levente Csák, F. Kuipers, B. Pace, David Hay, János Tapolcai
Guaranteeing a high availability of network services is a crucial part of network management. In this study, we show how to compute the availability of network services under earthquakes, by using empirical data. We take a multi-disciplinary approach and create an earthquake model based on seismological research and historical data. We then show how to integrate this empirical disaster model into existing network resiliency models to obtain the vulnerability and availability of a network under earthquakes. While previous studies have applied their models to ground shaking hazard models or earthquake scenarios, we compute (1) earthquake activity rates and (2) a relation between magnitude and disaster area, and use both as input data for our modeling. This approach is more in line with existing network resiliency models: it provides better information on the correlation between link failures than ground shaking hazard models and a more comprehensive view than a fixed set of scenarios.
{"title":"Network Resiliency Against Earthquakes","authors":"A. Valentini, Bálazs Vass, Jorik Oostenbrink, Levente Csák, F. Kuipers, B. Pace, David Hay, János Tapolcai","doi":"10.1109/RNDM48015.2019.8949088","DOIUrl":"https://doi.org/10.1109/RNDM48015.2019.8949088","url":null,"abstract":"Guaranteeing a high availability of network services is a crucial part of network management. In this study, we show how to compute the availability of network services under earthquakes, by using empirical data. We take a multi-disciplinary approach and create an earthquake model based on seismological research and historical data. We then show how to integrate this empirical disaster model into existing network resiliency models to obtain the vulnerability and availability of a network under earthquakes. While previous studies have applied their models to ground shaking hazard models or earthquake scenarios, we compute (1) earthquake activity rates and (2) a relation between magnitude and disaster area, and use both as input data for our modeling. This approach is more in line with existing network resiliency models: it provides better information on the correlation between link failures than ground shaking hazard models and a more comprehensive view than a fixed set of scenarios.","PeriodicalId":120852,"journal":{"name":"2019 11th International Workshop on Resilient Networks Design and Modeling (RNDM)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128985602","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}
Pub Date : 2019-10-01DOI: 10.1109/RNDM48015.2019.8949096
E. Calle, Sergio G. Cosgaya, D. Martinez, M. Pióro
Targeted attacks on software defined networks (SDN) can heavily disrupt the network connectivity and thus service continuity. One of the worst disruptions occurs when a targeted attack on network nodes is able to divide the network into different components (islands). Since the network is usually managed by a limited number of controllers, such attacks can result in isolating a significant proportion of nodes from the surviving controllers, causing major disruptions in service availability. In the paper we study the issue of adding a number of additional controllers to the network in order to increase its resilience to targeted attacks. For that we introduce an original optimization model for solving the related controller placement problem, based on a novel measure of network service availability called average network resilience (ANR). Additionally, we analyse two strategies to generate simultaneous targeted attack scenarios used in the numerical study illustrating the model effectiveness.
{"title":"Solving The Backup Controller Placement Problem In SDN Under Simultaneous Targeted Attacks","authors":"E. Calle, Sergio G. Cosgaya, D. Martinez, M. Pióro","doi":"10.1109/RNDM48015.2019.8949096","DOIUrl":"https://doi.org/10.1109/RNDM48015.2019.8949096","url":null,"abstract":"Targeted attacks on software defined networks (SDN) can heavily disrupt the network connectivity and thus service continuity. One of the worst disruptions occurs when a targeted attack on network nodes is able to divide the network into different components (islands). Since the network is usually managed by a limited number of controllers, such attacks can result in isolating a significant proportion of nodes from the surviving controllers, causing major disruptions in service availability. In the paper we study the issue of adding a number of additional controllers to the network in order to increase its resilience to targeted attacks. For that we introduce an original optimization model for solving the related controller placement problem, based on a novel measure of network service availability called average network resilience (ANR). Additionally, we analyse two strategies to generate simultaneous targeted attack scenarios used in the numerical study illustrating the model effectiveness.","PeriodicalId":120852,"journal":{"name":"2019 11th International Workshop on Resilient Networks Design and Modeling (RNDM)","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128458008","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}
Pub Date : 2019-10-01DOI: 10.1109/RNDM48015.2019.8949155
Marinela Shehaj, I. Kalesnikau, D. Nace, M. Pióro, Ermioni Qafzezi
This work deals with wireless mesh networks (WMN) composed of FSO (free space optics) links. Although FSO links realize broadband transmission at low cost, their drawback is sensitivity to adverse weather conditions causing transmission degradation on multiple links. Hence, designing such FSO networks requires an optimization model to find the cheapest configuration of link capacities that will be able to carry an acceptable level of the demanded traffic in all weather states that can be foreseen in network operation. Such a model can be achieved using robust optimization techniques, and for that it is important to find a tractable way of characterizing possible link (capacity) degradation states corresponding to weather conditions not known in advance. In the paper we show how weather records can be translated to vectors of link capacity available (after degradation) in particular weather states, and how the resulting link degradation states may be represented mathematically in a compact and tractable way to be exploited in optimization. To solve the second task we will make use of a generalization of a combinatorial problem of finding a minimum hitting set to deduce a compact set approximating a given set of link degradation states. Finally, we illustrate the effectiveness of our model by means of a numerical study.
{"title":"Modeling Transmission Degradation on FSO Links Caused by Weather Phenomena for WMN Optimization","authors":"Marinela Shehaj, I. Kalesnikau, D. Nace, M. Pióro, Ermioni Qafzezi","doi":"10.1109/RNDM48015.2019.8949155","DOIUrl":"https://doi.org/10.1109/RNDM48015.2019.8949155","url":null,"abstract":"This work deals with wireless mesh networks (WMN) composed of FSO (free space optics) links. Although FSO links realize broadband transmission at low cost, their drawback is sensitivity to adverse weather conditions causing transmission degradation on multiple links. Hence, designing such FSO networks requires an optimization model to find the cheapest configuration of link capacities that will be able to carry an acceptable level of the demanded traffic in all weather states that can be foreseen in network operation. Such a model can be achieved using robust optimization techniques, and for that it is important to find a tractable way of characterizing possible link (capacity) degradation states corresponding to weather conditions not known in advance. In the paper we show how weather records can be translated to vectors of link capacity available (after degradation) in particular weather states, and how the resulting link degradation states may be represented mathematically in a compact and tractable way to be exploited in optimization. To solve the second task we will make use of a generalization of a combinatorial problem of finding a minimum hitting set to deduce a compact set approximating a given set of link degradation states. Finally, we illustrate the effectiveness of our model by means of a numerical study.","PeriodicalId":120852,"journal":{"name":"2019 11th International Workshop on Resilient Networks Design and Modeling (RNDM)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129832417","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}
Pub Date : 2019-10-01DOI: 10.1109/RNDM48015.2019.8949149
Tamás Lévai, G. Rétvári
Programmable software switches have become crucial building blocks in a wide range of applications, from (virtual) data center networking to telco clouds. Modern use-cases, such as 5G, require very low latency, large throughput, and high availability. Software switches generally realize the packet processing pipeline as a dataflow graph: graph nodes correspond to simple packet processing operations and graph edges represent the control flow. The efficiency and dependability of the software switch critically depends on the way the dataflow graph is mapped to the underlying hardware resources. In this paper, we focus on dataflow graph embedding in a software switching context. We present an embedding approach which minimizes performance loss on inter-CPU communication across packet-processing control flow chains, and is resilient against a single CPU failure. The embedding is easy to generalize to $N$ CPU failures. We formulate the dataflow graph embedding problem as a mathematical program, characterize the computational complexity, and we propose optimal and heuristic algorithms to solve it. The viability of our approach is confirmed in comprehensive numerical analysis on a 5G packet processing pipeline, taken from an industrial 5G NFV benchmark suite.
{"title":"Resilient Dataflow Graph Embedding for Programmable Software Switches","authors":"Tamás Lévai, G. Rétvári","doi":"10.1109/RNDM48015.2019.8949149","DOIUrl":"https://doi.org/10.1109/RNDM48015.2019.8949149","url":null,"abstract":"Programmable software switches have become crucial building blocks in a wide range of applications, from (virtual) data center networking to telco clouds. Modern use-cases, such as 5G, require very low latency, large throughput, and high availability. Software switches generally realize the packet processing pipeline as a dataflow graph: graph nodes correspond to simple packet processing operations and graph edges represent the control flow. The efficiency and dependability of the software switch critically depends on the way the dataflow graph is mapped to the underlying hardware resources. In this paper, we focus on dataflow graph embedding in a software switching context. We present an embedding approach which minimizes performance loss on inter-CPU communication across packet-processing control flow chains, and is resilient against a single CPU failure. The embedding is easy to generalize to $N$ CPU failures. We formulate the dataflow graph embedding problem as a mathematical program, characterize the computational complexity, and we propose optimal and heuristic algorithms to solve it. The viability of our approach is confirmed in comprehensive numerical analysis on a 5G packet processing pipeline, taken from an industrial 5G NFV benchmark suite.","PeriodicalId":120852,"journal":{"name":"2019 11th International Workshop on Resilient Networks Design and Modeling (RNDM)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126104887","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：