Daphne Tuncer

Adaptive Resource Management and Control in Software Defined Networks

The heterogeneous nature of the applications, technologies and equipment that todays networks have to support has made the management of such infrastructures a complex task. The Software-Defined Networking (SDN) paradigm has emerged as a promising solution to reduce this complexity through the creation of a unified control plane independent of specific vendor equipment. However, designing a SDN-based solution for network resource management raises several challenges as it should exhibit flexibility, scalability and adaptability. In this paper, we present a new SDN-based management and control framework for fixed backbone networks, which provides support for both static and dynamic resource management applications. The framework consists of three layers which interact with each other through a set of interfaces. We develop a placement algorithm to determine the allocation of managers and controllers in the proposed distributed management and control layer. We then show how this layer can satisfy the requirements of two specific applications for adaptive load-balancing and energy management purposes.

More control over network resources: An ISP caching perspective

Management operations performed by Content Delivery Network (CDN) providers consist mainly in controlling the placement of contents at different storage locations and deciding where to serve client requests from. Configuration decisions are usually taken by using only limited information about the carrier networks, and this can adversely affect network usage. In this work we propose an approach by which ISPs can have more control over their resources. This involves the deployment of caching points within their network, which can allow them to implement their own content placement strategies. The work presented in this paper investigates lightweight strategies that can be used by the ISPs to manage the placement of contents in the various network caching locations according to user demand characteristics. The proposed strategies differ in terms of the volume and nature of the information required to determine the new caching configurations. We evaluate the performance of the proposed strategies, in terms of network resource utilization, based on a wide range of user demand profiles and we compare the obtained performance according to metrics we define to characterize the demand. The results demonstrate that the proposed metrics can provide useful indications regarding the performance one strategy can achieve over another and, as such, can be used by the ISP to improve the utilization of network resources.

DACoRM: A coordinated, decentralized and adaptive network resource management scheme

In order to meet the requirements of emerging demanding services, network resource management functionality that is decentralized, flexible and adaptive to traffic and network dynamics is of paramount importance. In this paper we describe the main mechanisms of DACoRM, a new intra-domain adaptive resource management approach for IP networks. Based on path diversity provided by multi-topology routing, our approach controls the distribution of traffic load in the network in an adaptive manner through periodical re-configurations that uses real-time monitoring information. The re-configuration actions performed are decided in a coordinated fashion between a set of source nodes that form an in-network overlay. We evaluate the overall performance of our approach using realistic network topologies. Results show that near-optimal network performance in terms of resource utilization can be achieved in scalable manner.

Managing the future internet through intelligent in-network substrates

The current Internet has been founded on the architectural premise of a simple network service used to interconnect relatively intelligent end systems. While this simplicity allowed it to reach an impressive scale, the predictive manner in which ISP networks are currently planned and configured through external management systems and the uniform treatment of all traffic are hampering its use as a unifying multi-service network. The future Internet will need to be more intelligent and adaptive, optimizing continuously the use of its resources and recovering from transient problems, faults and attacks without any impact on the demanding services and applications running over it. This article describes an architecture that allows intelligence to be introduced within the network to support sophisticated self-management functionality in a coordinated and controllable manner. The presented approach, based on intelligent substrates, can potentially make the Internet more adaptable, agile, sustainable, and dependable given the requirements of emerging services with highly demanding traffic and rapidly changing locations. We discuss how the proposed framework can be applied to three representative emerging scenarios: dynamic traffic engineering (load balancing across multiple paths); energy efficiency in ISP network infrastructures; and cache management in content-centric networks.

Scalable Cache Management for ISP-Operated Content Delivery Services

Content delivery networks (CDNs) have been the prevalent method for the efficient delivery of content across the Internet. Management operations performed by CDNs are usually applied only based on limited information about Internet Service Provider (ISP) networks, which can have a negative impact on the utilization of ISP resources. To overcome these issues, previous research efforts have been investigating ISP-operated content delivery services, by which an ISP can deploy its own in-network caching infrastructure and implement its own cache management strategies. In this paper, we extend our previous work on ISP-operated content distribution and develop a novel scalable and efficient distributed approach to control the placement of content in the available caching points. The proposed approach relies on parallelizing the decision-making process and the use of network partitioning to cluster the distributed decision-making points, which enables fast reconfiguration and limits the volume of information required to take reconfiguration decisions. We evaluate the performance of our approach based on a wide range of parameters. The results demonstrate that the proposed solution can outperform previous approaches in terms of management overhead and complexity while offering similar network and caching performance.

Proactive multi-tenant cache management for virtualized ISP networks

The content delivery market has mainly been dominated by large Content Delivery Networks (CDNs) such as Akamai and Limelight. However, CDN traffic exerts a lot of pressure on Internet Service Provider (ISP) networks. Recently, ISPs have begun deploying so-called Telco CDNs, which have many advantages, such as reduced ISP network bandwidth utilization and improved Quality of Service (QoS) by bringing content closer to the end-user. Virtualization of storage and networking resources can enable the ISP to simultaneously lease its Telco CDN infrastructure to multiple third parties, opening up new business models and revenue streams. In this paper, we propose a proactive cache management system for ISP-operated multitenant Telco CDNs. The associated algorithm optimizes content placement and server selection across tenants and users, based on predicted content popularity and the geographical distribution of requests. Based on a Video-on-Demand (VoD) request trace of a leading European telecom operator, the presented algorithm is shown to reduce bandwidth usage by 17% compared to the traditional Least Recently Used (LRU) caching strategy, both inside the network and on the ingress links, while at the same time offering enhanced load balancing capabilities. Increasing the prediction accuracy is shown to have the potential to further improve bandwidth efficiency by up to 79%.

A hybrid management substrate structure for adaptive network resource management

Centralized and offline network management functionality, traditionally deployed by operators, cannot easily deal with the traffic patterns of emerging services, which are becoming more dynamic and unpredictable. As such, decentralized solutions that are flexible and adaptive to traffic and network dynamics are of paramount importance. To this end, we have been developing an in-network management approach in which an intelligent substrate allows the dynamic reconfiguration of resources according to network conditions. The set of nodes forming this logical structure are able to communicate with each other to coordinate their decisions. While in previous work we investigated the use of full-mesh and ring structures to connect the substrate nodes, we consider here a hybrid approach that combines the benefits of the other two. We describe algorithms that can be practically used to compute this hybrid structure and that take into account important criteria such as minimizing the latency and the communication overhead among the substrate nodes. We evaluate the impact of key parameters associated with the construction process.

Towards multi-tenant cache management for ISP networks

The decreasing cost of storage and the advent of virtualization technology can allow Internet Service Providers (ISPs) to deploy multi-tenant caching infrastructures and lease them to content producers and Content Delivery Networks (CDNs). Serving content requests directly from the ISP network does not only reduce the delivery time, but also allows the ISP to optimize the network resources by controlling the placement and routing of content items. In this paper, we introduce a multi-tenant cache management approach that significantly reduces the bandwidth utilization of ISPs networks by pro-actively allocating caching space, leased by content producers and/or CDNs, and intelligently routing content to the end users. Using real content request traces, we show that the optimal solution to this problem can increase the cache hit ratio by 70.64% while reducing the bandwidth usage by 57.17% on average, compared to a commonly used reactive cache management scheme. These results provide a benchmark for the development of novel multi-tenant cache management strategies.

Extensible signaling framework for decentralized network management applications

The management of network infrastructures has become increasingly complex over time, which is mainly attributed to the introduction of new functionality to support emerging services and applications. To address this important issue, research efforts in the last few years focused on developing Software-Defined Networking solutions. While initial work proposed centralized architectures, their scalability limitations have led researchers to investigate a distributed control plane, with controller placement algorithms and mechanisms for building a logically centralized network view, being examples of challenges addressed. A critical issue that has not been adequately addressed concerns the communication between distributed decision-making entities to ensure configuration consistency. To this end, this paper proposes a signaling framework that can allow the exchange of information in distributed management and control scenarios. The benefits of the proposed framework are illustrated through a realistic network resource management use case. Based on simulation, we demonstrate the flexibility and extensibility of our solution in meeting the requirements of distributed decision-making processes.

Hybrid multi-tenant cache management for virtualized ISP networks

In recent years, Internet Service Providers (ISPs) have started to deploy Telco Content Delivery Networks (Telco CDNs) to reduce the pressure on their network resources. The deployment of Telco CDNs results in reduced ISP bandwidth utilization and improved service quality by bringing the content closer to the end-users. Furthermore, virtualization of storage and networking resources can open up new business models by enabling the ISP to simultaneously lease its Telco CDN infrastructure to multiple third parties. Previous work has shown that multi-tenant proactive resource allocation and content placement can significantly reduce the load on the ISP network. However, the performance of this approach strongly depends on the prediction accuracy for future content requests. In this paper, a hybrid cache management approach is proposed where proactive content placement and traditional reactive caching strategies are combined. In this way, content placement and server selection can be optimized across tenants and users, based on predicted content popularity and the geographical distribution of requests, while simultaneously providing reactivity to unexpected changes in the request pattern. Based on a Video-on-Demand (VoD) production request trace, it is shown that the total hit ratio can be increased by 43% while using 5% less bandwidth compared to the traditional Least Recently Used (LRU) caching strategy. Furthermore, the proposed approach requires 39% less migration overhead compared to the proactive placement approach we previously proposed in Claeys et al. (2014b) and achieves a hit ratio increase of 19% and bandwidth usage reduction of 7% in the evaluated VoD scenarios and topology.