2018

• Supervisor: Loh, F.: Evaluation of the Video Streaming Behavior Using Amazon Echo Show, (2018).
  [ BibTeX ] [ EndNote ]
   
  @mastersthesis{info3-bachelor-2018-6, author = {Supervisor: Loh, Frank}, school = {University of Würzburg}, title = {Evaluation of the Video Streaming Behavior Using Amazon Echo Show}, year = 2018 }
  %0 Thesis %1 info3-bachelor-2018-6 %A Supervisor: Loh, Frank %D 2018 %T Evaluation of the Video Streaming Behavior Using Amazon Echo Show
• Supervisor: Grigorjew, A.: Consolidation of Resources and Demands in VNF Chain Placement Problems (BA), (2018).
  [ Abstract ] [ BibTeX ] [ EndNote ]
   
  Contact: Alexej Grigorjew <alexej.grigorjew@informatik.uni-wuerzburg.de> Placing Virtual Network Functions in network graphs is an NP-hard problem. Solving large problem instances is often unfeasible due to limited computational resources and time. For many use cases, the problem input is not required in all detail, but can be aggregated into a smaller abstraction. In this work, an algorithm should be implemented that generates such smaller representations of graphs and VNF demands. Further, its performance shall be evaluated by applying (existing) solvers on the results and comparing both runtime and quality of their placements with the original graph.
  @mastersthesis{info3-bachelor-2017-30, abstract = {Contact: Alexej Grigorjew <alexej.grigorjew@informatik.uni-wuerzburg.de> Placing Virtual Network Functions in network graphs is an NP-hard problem. Solving large problem instances is often unfeasible due to limited computational resources and time. For many use cases, the problem input is not required in all detail, but can be aggregated into a smaller abstraction. In this work, an algorithm should be implemented that generates such smaller representations of graphs and VNF demands. Further, its performance shall be evaluated by applying (existing) solvers on the results and comparing both runtime and quality of their placements with the original graph.}, author = {Supervisor: Grigorjew, Alexej}, school = {University of Würzburg}, title = {Consolidation of Resources and Demands in VNF Chain Placement Problems (BA)}, year = 2018 }
  %0 Thesis %1 info3-bachelor-2017-30 %A Supervisor: Grigorjew, Alexej %D 2018 %T Consolidation of Resources and Demands in VNF Chain Placement Problems (BA) %X Contact: Alexej Grigorjew <alexej.grigorjew@informatik.uni-wuerzburg.de> Placing Virtual Network Functions in network graphs is an NP-hard problem. Solving large problem instances is often unfeasible due to limited computational resources and time. For many use cases, the problem input is not required in all detail, but can be aggregated into a smaller abstraction. In this work, an algorithm should be implemented that generates such smaller representations of graphs and VNF demands. Further, its performance shall be evaluated by applying (existing) solvers on the results and comparing both runtime and quality of their placements with the original graph.
• Gray, N.: Evaluation of the Impact of Network Topologies and Applications on Synchronization Strategies of Distributed SDN Controllers, (2018).
  [ Abstract ] [ BibTeX ] [ EndNote ]
   
  Software-defined Networking (SDN) separates the data from the control plane and outsources the latter to a logically centralized instance called the controller. This enables a common view of the network topology and provides a central entity for network configurations. The controller is often distributed to impose no single point of failure and to balance the load efficiently among the physical controller instances. To maintain a consistent view of the network, synchronization between the individual controller instances is required. Controller implementations can be categorized by their utilized synchronization strategy and largely fall in one of two categories, namely horizontal or hierarchical. Goal of this thesis is to evaluate the impact of the network topology and running controller applications on the performance of the controller when using different synchronization mechanisms. As access to large SDN test beds is limited the investigations will be done by simulations within the OMNeT++ framework.
  @mastersthesis{info3-bachelor-2017-18, abstract = {Software-defined Networking (SDN) separates the data from the control plane and outsources the latter to a logically centralized instance called the controller. This enables a common view of the network topology and provides a central entity for network configurations. The controller is often distributed to impose no single point of failure and to balance the load efficiently among the physical controller instances. To maintain a consistent view of the network, synchronization between the individual controller instances is required. Controller implementations can be categorized by their utilized synchronization strategy and largely fall in one of two categories, namely horizontal or hierarchical. Goal of this thesis is to evaluate the impact of the network topology and running controller applications on the performance of the controller when using different synchronization mechanisms. As access to large SDN test beds is limited the investigations will be done by simulations within the OMNeT++ framework.}, author = {Gray, Nicholas}, school = {University of Würzburg}, title = {Evaluation of the Impact of Network Topologies and Applications on Synchronization Strategies of Distributed SDN Controllers}, year = 2018 }
  %0 Thesis %1 info3-bachelor-2017-18 %A Gray, Nicholas %D 2018 %T Evaluation of the Impact of Network Topologies and Applications on Synchronization Strategies of Distributed SDN Controllers %X Software-defined Networking (SDN) separates the data from the control plane and outsources the latter to a logically centralized instance called the controller. This enables a common view of the network topology and provides a central entity for network configurations. The controller is often distributed to impose no single point of failure and to balance the load efficiently among the physical controller instances. To maintain a consistent view of the network, synchronization between the individual controller instances is required. Controller implementations can be categorized by their utilized synchronization strategy and largely fall in one of two categories, namely horizontal or hierarchical. Goal of this thesis is to evaluate the impact of the network topology and running controller applications on the performance of the controller when using different synchronization mechanisms. As access to large SDN test beds is limited the investigations will be done by simulations within the OMNeT++ framework.

2018

• Supervisor: Geissler, S.: Investigating the Feasibility of Machine Learning for Placement Problems in Softwarized Networks, (2018).
  [ Abstract ] [ BibTeX ] [ EndNote ]
   
  Softwarization is a crucial part of modern and next generation networks, increasing flexibility, scalability and efficiency. However, new challenges regarding the placement of entities like SDN controllers or virtual network functions need to be addressed in order to reap those benefits. Unfortunately, most of these problems are NP-hard and can thus not be solved in an efficient way. In order to alleviate this issue, the goal of this thesis is to investigate the application of machine learning approaches to a special version of a placement problem, namely SDN controller placement. Since this problem can be solved extensively in a reasonably time frame, it is possible to evaluate the machine learning approach to a common ground truth and generate a reasonable amount of training data for the machine learning model.
  @mastersthesis{info3-master-2017-15, abstract = {Softwarization is a crucial part of modern and next generation networks, increasing flexibility, scalability and efficiency. However, new challenges regarding the placement of entities like SDN controllers or virtual network functions need to be addressed in order to reap those benefits. Unfortunately, most of these problems are NP-hard and can thus not be solved in an efficient way. In order to alleviate this issue, the goal of this thesis is to investigate the application of machine learning approaches to a special version of a placement problem, namely SDN controller placement. Since this problem can be solved extensively in a reasonably time frame, it is possible to evaluate the machine learning approach to a common ground truth and generate a reasonable amount of training data for the machine learning model.}, author = {Supervisor: Geissler, Stefan}, school = {University of Wuerzburg}, title = {Investigating the Feasibility of Machine Learning for Placement Problems in Softwarized Networks}, year = 2018 }
  %0 Thesis %1 info3-master-2017-15 %A Supervisor: Geissler, Stefan %D 2018 %T Investigating the Feasibility of Machine Learning for Placement Problems in Softwarized Networks %X Softwarization is a crucial part of modern and next generation networks, increasing flexibility, scalability and efficiency. However, new challenges regarding the placement of entities like SDN controllers or virtual network functions need to be addressed in order to reap those benefits. Unfortunately, most of these problems are NP-hard and can thus not be solved in an efficient way. In order to alleviate this issue, the goal of this thesis is to investigate the application of machine learning approaches to a special version of a placement problem, namely SDN controller placement. Since this problem can be solved extensively in a reasonably time frame, it is possible to evaluate the machine learning approach to a common ground truth and generate a reasonable amount of training data for the machine learning model.
• Supervisor: Grigorjew, A.: Development and Evaluation of Algorithms for Online Distributed Virtual Network Function Deployment and Consolidation (MA), (2018).
  [ Abstract ] [ BibTeX ] [ EndNote ]
   
  Contact: Alexej Grigorjew <alexej.grigorjew@informatik.uni-wuerzburg.de> Network Function Virtualization (NFV) is an emerging trend in the field of network design and operation. Traditionally, various network functions (such as Firewalls, Intrusion Detection Systems, Deep Packet Inspection, ...) are applied to flows in the network, and currently implemented by specialized hardware middleboxes. In the context of NFV, these network functions are now implemented in software and deployed on virtual machines running on commodity servers (→ Virtual Network Functions, VNFs). This enables far greater flexibility and, consequently, comes with new optimization problems. One important part of these problems is to decide how many of these VNF instances are needed, and where they should be placed. Together with various constraints (bandwidth, CPU capacities, ...) and objectives (minimize costs, delays, ...), these questions form an NP-hard optimization problem. In this particular work, the online version of this problem shall be investigated, i.e., new instances are created and removed from the network during its operation, as new traffic demands arrive or leave the network. The aim is to design algorithms that are able to make these decisions locally (i.e., no central decision unit) and to evaluate their performance with respect to chosen objectives, for example by means of a simulation. The student shall (among other things) ... - formalize the problem and its context, - outline its relation to existing, related approaches, - develop optimization algorithms for VNF placement and removal, - define appropriate measures for the quality of attained placements, - evaluate the placement quality by means of these measures (→ discrete event simulation), - and interpret attained results. For further information, feel free to contact us in person or send an email.
  @mastersthesis{info3-master-2017-9, abstract = {Contact: Alexej Grigorjew <alexej.grigorjew@informatik.uni-wuerzburg.de> Network Function Virtualization (NFV) is an emerging trend in the field of network design and operation. Traditionally, various network functions (such as Firewalls, Intrusion Detection Systems, Deep Packet Inspection, ...) are applied to flows in the network, and currently implemented by specialized hardware middleboxes. In the context of NFV, these network functions are now implemented in software and deployed on virtual machines running on commodity servers (→ Virtual Network Functions, VNFs). This enables far greater flexibility and, consequently, comes with new optimization problems. One important part of these problems is to decide how many of these VNF instances are needed, and where they should be placed. Together with various constraints (bandwidth, CPU capacities, ...) and objectives (minimize costs, delays, ...), these questions form an NP-hard optimization problem. In this particular work, the online version of this problem shall be investigated, i.e., new instances are created and removed from the network during its operation, as new traffic demands arrive or leave the network. The aim is to design algorithms that are able to make these decisions locally (i.e., no central decision unit) and to evaluate their performance with respect to chosen objectives, for example by means of a simulation. The student shall (among other things) ... - formalize the problem and its context, - outline its relation to existing, related approaches, - develop optimization algorithms for VNF placement and removal, - define appropriate measures for the quality of attained placements, - evaluate the placement quality by means of these measures (→ discrete event simulation), - and interpret attained results. For further information, feel free to contact us in person or send an email.}, author = {Supervisor: Grigorjew, Alexej}, school = {University of Würzburg}, title = {Development and Evaluation of Algorithms for Online Distributed Virtual Network Function Deployment and Consolidation (MA)}, year = 2018 }
  %0 Thesis %1 info3-master-2017-9 %A Supervisor: Grigorjew, Alexej %D 2018 %T Development and Evaluation of Algorithms for Online Distributed Virtual Network Function Deployment and Consolidation (MA) %X Contact: Alexej Grigorjew <alexej.grigorjew@informatik.uni-wuerzburg.de> Network Function Virtualization (NFV) is an emerging trend in the field of network design and operation. Traditionally, various network functions (such as Firewalls, Intrusion Detection Systems, Deep Packet Inspection, ...) are applied to flows in the network, and currently implemented by specialized hardware middleboxes. In the context of NFV, these network functions are now implemented in software and deployed on virtual machines running on commodity servers (→ Virtual Network Functions, VNFs). This enables far greater flexibility and, consequently, comes with new optimization problems. One important part of these problems is to decide how many of these VNF instances are needed, and where they should be placed. Together with various constraints (bandwidth, CPU capacities, ...) and objectives (minimize costs, delays, ...), these questions form an NP-hard optimization problem. In this particular work, the online version of this problem shall be investigated, i.e., new instances are created and removed from the network during its operation, as new traffic demands arrive or leave the network. The aim is to design algorithms that are able to make these decisions locally (i.e., no central decision unit) and to evaluate their performance with respect to chosen objectives, for example by means of a simulation. The student shall (among other things) ... - formalize the problem and its context, - outline its relation to existing, related approaches, - develop optimization algorithms for VNF placement and removal, - define appropriate measures for the quality of attained placements, - evaluate the placement quality by means of these measures (→ discrete event simulation), - and interpret attained results. For further information, feel free to contact us in person or send an email.
• Gray, N.: Evaluation of Offloading Firewall Rules with P4, (2018).
  [ Abstract ] [ BibTeX ] [ EndNote ]
   
  Contact: nicholas.gray@informatik.uni-wuerzburg.de Firewalls are classical network middle boxes, which are physically implemented within the data path. Following this design pattern causes the network to become inflexible and harder to scale as additional hardware devices have to be deployed whenever the limits of the current configuration are reached, resulting in a high maintenance and requisition cost. Network Function Virtualization (NFV) is a novel paradigm aiming to mitigate these drawbacks by shifting the function of hardware middle boxes to software programs run on Commodity of the Shelf servers. While increasing the network flexibility, it also imposes additional delays on the data path as the complete networking and software stack has to be traversed for each packet. P4 allows the execution and packet matching process on inexpensive, programmable networking cards at line rate. Goal of the thesis is the implementation and evaluation of a distributed stateful firewall based on P4.
  @mastersthesis{info3-master-2017-7, abstract = {Contact: nicholas.gray@informatik.uni-wuerzburg.de Firewalls are classical network middle boxes, which are physically implemented within the data path. Following this design pattern causes the network to become inflexible and harder to scale as additional hardware devices have to be deployed whenever the limits of the current configuration are reached, resulting in a high maintenance and requisition cost. Network Function Virtualization (NFV) is a novel paradigm aiming to mitigate these drawbacks by shifting the function of hardware middle boxes to software programs run on Commodity of the Shelf servers. While increasing the network flexibility, it also imposes additional delays on the data path as the complete networking and software stack has to be traversed for each packet. P4 allows the execution and packet matching process on inexpensive, programmable networking cards at line rate. Goal of the thesis is the implementation and evaluation of a distributed stateful firewall based on P4.}, author = {Gray, Nicholas}, school = {University of Wuerzburg}, title = {Evaluation of Offloading Firewall Rules with P4}, year = 2018 }
  %0 Thesis %1 info3-master-2017-7 %A Gray, Nicholas %D 2018 %T Evaluation of Offloading Firewall Rules with P4 %X Contact: nicholas.gray@informatik.uni-wuerzburg.de Firewalls are classical network middle boxes, which are physically implemented within the data path. Following this design pattern causes the network to become inflexible and harder to scale as additional hardware devices have to be deployed whenever the limits of the current configuration are reached, resulting in a high maintenance and requisition cost. Network Function Virtualization (NFV) is a novel paradigm aiming to mitigate these drawbacks by shifting the function of hardware middle boxes to software programs run on Commodity of the Shelf servers. While increasing the network flexibility, it also imposes additional delays on the data path as the complete networking and software stack has to be traversed for each packet. P4 allows the execution and packet matching process on inexpensive, programmable networking cards at line rate. Goal of the thesis is the implementation and evaluation of a distributed stateful firewall based on P4.

2018

• Supervisor: Wamser, F.: Implementing a Load Indicator Metric for Network Functions, (2018).
   
• Supervisor: Grigorjew, A.: Performance and Parameter Evaluation of P4 Devices (Prakt.), (2018).
  [ Abstract ]
   
  Contact: Alexej Grigorjew <alexej.grigorjew@informatik.uni-wuerzburg.de> P4 is a simple programming language for switching devices that allows free configuration of protocols, matching fields, actions, and control flow in compatible hardware and software implementations. This leads to a wide range of different configurations and different realizations of their behavior by different vendors, coupled with tight performance requirements in the data plane. In this work, the influence of various parameters on the performance of P4 implementations shall be measured. In particular, a software implementation and a Netronome P4 hardware NIC shall be compared and evaluated with regard to their program's impact on measures such as throughput and processing delay.
• Supervisor: Borchert, K.: Analysis and Correlation of objective and subjective Performance Data of an Enterprise Application, (2018).
  [ Abstract ]
   
  Contact: kathrin.borchert@informatik.uni-wuerzburg.de
• Supervisor: Wamser, F.: Aufbau eines Testbeds zur Demonstration von dynamischen Diensten in der Cloud, (2018).
  [ Abstract ]
   
  Cloud Service Placement for Live Video Streaming with a Remote-Controlled Drone florian.wamser@informatik.uni-wuerzburg.de