Rhaban Hark

DYNSDM: Dynamic and flexible software-defined multicast for ISP environments

A number of todays over-the-top (OTT) services could greatly benefit from a scalable and efficient network-layer multicast support on the Internet. IP multicast showed to not meet these requirements and, thus, is not available for this purpose. Content Delivery Networks emerged as global alternative but usually end at the border of ISP networks. Software-Defined Multicast (SDM) is proposed in a previous work by the authors, enabling ISP-internal network-layer multicast delivery of OTT traffic. While it coins fundamental concepts, it does not detail the ISP-internal traffic and service management and leaves important questions unanswered. To this end, DYNSDM is proposed in this paper, detailing the multicast planning and management, proposing a novel network-layer multi-tree mechanism to distribute traffic on links inside the ISP network, and introducing mechanisms to handle group and network dynamics. DYNSDM was prototypically evaluated, showing its high traffic efficiency, good scalability, and superior traffic distribution characteristics.

Flexible, Efficient, and Scalable Software-Defined Over-the-Top Multicast for ISP Environments With DynSdm

A number of todays over-the-top (OTT) services could greatly benefit from a flexible, efficient, and scalable network-layer multicast support on the Internet. IP multicast showed not to meet these requirements, hindering its adoption in more than individual network islands. As a consequence, content delivery networks (CDNs) emerged as alternative and implemented one-to-many multicasting at the application layer. Yet, due to practical reasons and conflicting stakeholder interests, CDN deployments usually end at the border of Internet service provider networks, making one-to-many deliveries at an Internet-scale challenging up to date. In a previous work, the authors presented software-defined multicast (Sdm) to mitigate this problem by unlocking network-layer multicast support for OTT services inside ISP networks. In this paper, DYNSDM is proposed to complement Sdm in key aspects, allowing for a more holistic support of relevant scenarios aspects, and enabling a fine-granular traffic engineering of multicast traffic inside ISP networks. For this, DYNSDM presents a detailed design of the ISP-internal traffic and service management process and introduces a set of novel SDN-based network-layer mechanisms for traffic load balancing as well as handling of group and network dynamics. DYNSDM was implemented as prototype and extensively evaluated, showing its flexibility, high traffic efficiency, good scalability, and superior ability to balance traffic load inside ISP networks.

Stateless Gateways - Reducing Cellular Traffic for Event Distribution in Mobile Social Applications

The increasing capabilities of modern smartphones lead to the design of mobile social applications focusing on direct interaction between users. Ranging from mobile social networks to fully-fledged augmented reality games, these applications usually operate on contextual information to identify relevant content - most notably, the users physical location. The resulting locality properties of the interaction between users are not well reflected by the cloud-based, centralized infrastructure utilized in todays mobile applications. Thus, data that is relevant to a group of nearby users is downloaded multiple times via the cellular network. Due to capped and/or expensive data tariffs, this can have severe impact on the user acceptance of such applications. To address this issue, we propose the concept of stateless gateways to augment cloud-based mobile social applications. A stateless gateway is chosen by the cloud to distribute information to nearby interested parties, without requiring any additional state information on the gateway itself. We integrate the concept into a location-based publish/subscribe system and show the resulting performance and cost characteristics through extensive evaluations. Our results show that the stateless concept enables frequent gateway switches, lowering the load on the cellular network by 70% for the scenario of a mobile augmented reality game. At the same time, our system achieves better fairness characteristics among participants due to a more efficient utilization of gateway nodes compared to a less flexible assignment of gateways.

A Framework for Publish/Subscribe Protocol Transitions in Mobile Crowds

The increasing number of sensor-equipped mobile devices enables new applications for communication within crowds of people. Ranging from monitoring services that provide insights into the crowds behavior to fully fledged messaging applications for end users, all such applications require communication protocols that are tailored to the characteristics of a mobile ad hoc network MANET. A common communication scheme for such dynamic networks is the publish/subscribe pub/sub paradigm, that offers temporal and spatial decoupling. However, pub/sub protocols for MANETs are designed and evaluated under very restricted conditions regarding the expected mobility, number of users, application workload, and application requirements, targeting rather specific scenarios. In reality, such conditions are subject to change, especially considering peoples behavior during crowded events. In this work, we propose a framework that enables dynamic transitions between different pub/sub protocols based on the currently monitored conditions or caused by an external trigger. We analyze the behavior during transitions and the overhead introduced by our framework through extensive simulations. By using simple pub/sub protocols and switching between them based on the observed conditions, we are able to maintain good service quality at reasonable overhead, thereby enabling applications to operate in dynamic conditions representative of mobile crowds.

DistTM: Collaborative traffic matrix estimation in distributed SDN control planes

Recently, several works propose monitoring approaches for the emerged paradigm of Software-defined Networking. These provide a couple of ideas to retrieve various information about the network state leveraging new concepts for monitoring data collection at flow-level. As existing approaches reduce their scope to networks with a single controller, even sophisticated approaches ignore a potentially great efficiency gap, due to redundant flow measurements by multiple controllers in adjacent networks. To show a possibility how to close this efficiency gap, we propose a solution for collaborative traffic matrix estimation, termed DistTM. It exploits the property that flows traverse multiple networks and are monitored by several controllers. Through collaboration, the resulting monitoring tasks are coordinated and distributed among participating controllers to capture relevant information about all traversing flows, omitting redundant data collection. Conducted simulations reveal that DistTM operates efficiently: the monitoring traffic is significantly reduced, while the traffic matrix entry staleness is slightly affected. Furthermore, DistTM provides different schemes for a fair load balancing on controllers and switches while taking different influencing aspects into consideration.

Limiting the Footprint of Monitoring in Dynamic Scenarios through Multi-Dimensional Offloading

Multiple offloading techniques are used in todays communication systems. Most approaches rely on cellular offloading to reduce the burden on the cellular infrastructure, especially in crowded situations. While the load on the cellular infrastructure decreases, network participants that are actively involved in the networking effort by device-to-device communication carry the load of a majority of devices. Though, by focusing on the cellular medium for offloading, the increased number of public available network access points remains mostly unused by most approaches. Incorporating access points for offloading entails numerous advantages, such as a reduced load on the cellular data plan of mobile users. In this work, the potentials of multi-dimensional offloading are assessed using the example of a state-of-the-art adaptive monitoring system that, so far, only employs cellular and device-to-device offloading. We show that the system can benefit from adding additional components and protocols that enable multi-dimensional offloading. Through an extensive simulation study we show that combining different offloading techniques leads to significant improvements regarding the achieved service quality (up to 15-fold) and the responsiveness (up to 5-fold), while reducing the load on the mobile nodes by at least 35% even in scenarios with good cellular connectivity.

Towards Optimal Placement of Monitoring Units in Time-varying Networks under Centralized Control

The increasing penetration of software-defined communication networks with centralized control has made network management a highly demanding task. Common monitoring approaches in the context of such convoluted high-speed networks have become a serious challenge in terms of complexity and resource management. Management functions rely on monitoring information such as the flow size distribution (FSD), to perform crucial activities such as load balancing and resource provisioning. In this paper, we propose a solution as to how one can utilize limited monitoring resources to estimate the FSD for distinct flows characterized by origin-destination pairs. We provide a method to dynamically adapt placement of monitoring units with some extracted knowledge about the change in FSD’s with time.