Tim Wauters

Replica placement in ring based content delivery networks

The recent introduction of Content Distribution Networks (CDNs) enhances the delivery of high quality multimedia content to end users. In a CDN architecture, the content is replicated to so-called surrogate servers, generally at the edge of the transport network, to improve the quality of service (QoS) of streaming multimedia delivery services. By using peer-to-peer (P2P) technologies, these edge servers can co-operate and provide a more scalable and robust service in a self-organizing CDN. In this paper, we propose a set of distributed replica placement algorithms (RPAs), based on an Integer Linear Programming (ILP) formulation of the centralized content placement problem. These algorithms further enhance the CDN performance by optimizing the network and server load, reducing network delays and avoiding congestion. Although the proposed algorithms are designed for and tested on different network topologies, we focus on robust ring based CDNs in this study. Content placement on such a network topology can be calculated analytically and can be used for comparison.

Co-operative Proxy Caching Algorithms for Time-Shifted IPTV Services

The increasing popularity of multimedia streaming applications introduces new challenges in content distribution networks. Streaming services such as video on demand (VoD) or digital television over the Internet (IPTV) are very bandwidth-intensive and cannot tolerate the high delays and poor loss properties of todays Internet. To solve these problems, caching (a sliding segment of popular streams at proxies could be envisaged. This paper presents a novel caching algorithm and architecture for time-shifted television (tsTV) and its implementation using the IETFs real-time streaming protocol (RTSP). The algorithm uses sliding caching windows with sizes depending on content popularity and/or distance metrics. The caches can work in stand-alone mode as well as in co-operative mode. This paper shows that the network load can already be reduced considerably using small diskless caches, especially when using cooperative caching. A brief overview of the functionality of a prototype proxy implementation is presented as well

Towards a predictive cache replacement strategy for multimedia content

In recent years, telecom operators have been moving away from traditional broadcast-driven television, towards IP-based interactive and on-demand multimedia services. Consequently, multicast is no longer sufficient to limit the amount of generated traffic in the network. In order to prevent an explosive growth in traffic, caches can be strategically placed throughout the content delivery infrastructure. As the size of caches is usually limited to only a small fraction of the total size of all content items, it is important to accurately predict future content popularity. Traditional caching strategies only take into account the past when deciding what content to cache. Recently, a trend towards novel strategies that actually try to predict future content popularity has arisen. In this paper, we ascertain the viability of using popularity prediction in realistic multimedia content caching scenarios. The proposed generic popularity prediction algorithm is capable of predicting future content popularity, independent of specific content and service characteristics. Additionally, a novel cache replacement strategy, which employs the popularity prediction algorithm when making its decisions, is introduced. A detailed evaluation, based on simulation results using trace files from an actual deployed Video on Demand service, was performed. The evaluation results are used to determine the merits of popularity-based caching compared to traditional strategies. Additionally, the synergy between several parameters, such as cache size and prediction window, is investigated. Results show that the proposed prediction-based caching strategy has the potential to significantly outperform state-of-the-art traditional strategies. Specifically, the evaluated Video on Demand scenario showed a performance increase of up to 20% in terms of cache hit rate.

Self-optimized cognitive network of networks

Future processing, storage and communication services will be highly pervasive: people, smart objects, machines and the surrounding space (all embedding devices such as with sensors, RFID tags, etc.) will define a highly decentralized cyber environment of resources interconnected by dynamic Networks of Networks. As communications will extend to cover any combination of “people, machines and things”, future networks will be increasingly complex and heterogeneous, yet always endorsed with the challenging task of ensuring end-to-end QoS. This paper proposes groundwork for an advanced cognitive networking paradigm exploitable in future wired and wireless infrastructures: a Decentralised Cognitive Plane to allow for cross-layer, cross-node and cross-network domain self-management, self-control and self-optimization, whilst being compatible with legacy management and control.

A latency-aware algorithm for dynamic service placement in large-scale overlays

A generic and self-managing service hosting infrastructure, provides a means to offer a large variety of services to users across the Internet. Such an infrastructure provides mechanisms to automatically allocate resources to services, discover the location of these services, and route client requests to a suitable service instance.

HTTP/2-Based Methods to Improve the Live Experience of Adaptive Streaming

HTTP Adaptive Streaming (HAS) is today the number one video technology for over-the-top video distribution. In HAS, video content is temporally divided into multiple segments and encoded at different quality levels. A client selects and retrieves per segment the most suited quality version to create a seamless playout. Despite the ability of HAS to deal with changing network conditions, HAS-based live streaming often suffers from freezes in the playout due to buffer under-run, low average quality, large camera-to-display delay, and large initial/channel-change delay. Recently, IETF has standardized HTTP/2, a new version of the HTTP protocol that provides new features for reducing the page load time in Web browsing. In this paper, we present ten novel HTTP/2-based methods to improve the quality of experience of HAS. Our main contribution is the design and evaluation of a push-based approach for live streaming in which super-short segments are pushed from server to client as soon as they become available. We show that with an RTT of 300 ms, this approach can reduce the average server-to-display delay by 90.1% and the average start-up delay by 40.1%.

On the merits of popularity prediction in multimedia content caching

In recent years, telecom operators have been moving away from traditional, broadcast-driven, television towards IP-based, interactive and on-demand services. Consequently, multicast is no longer a viable solution to limit the amount of traffic in the IP-TV network. In order to counter an explosion in generated traffic, caches can be strategically placed throughout the content delivery infrastructure. As the size of caches is usually limited to only a small fraction of the total size of all content items, it is important to accurately predict future content popularity. Classical caching strategies only take into account the past when deciding what content to cache. Recently, a trend towards novel strategies that actually try to predict future content popularity has arisen. In this paper, we ascertain the viability of using popularity prediction in realistic multimedia content caching scenarios. The use of popularity prediction is compared to classical strategies using trace files from an actual deployed Video on Demand service. Additionally, the synergy between several parameters, such as cache size and prediction window, is investigated.

Optical network design for video on demand services

In this paper a network design for video on demand (VoD) services on Ethernet-based WDM networks is presented. The decentralised architecture consists of independent regional ring networks with local video servers. Based on an Integer Linear Programming (ILP) model, a network design tool, minimising the total installation cost on the core and the Hybrid Fiber Coax (HFC) access network, has been developed. Unicast as well as broadcast VoD services are taken into account. The influence of different parameters in our traffic and content models on the network design is studied.

Load balancing through efficient distributed content placement

The concept of content distribution networks (CDNs) has recently been introduced to enhance the delivery of bandwidth-intensive multimedia content to end users. In a CDN architecture, the content is replicated from the origin server to so-called surrogate servers at the edge of the Internet, to improve the quality of service and optimise network bandwidth usage. The introduction of peer-to-peer (P2P) architectures, where all nodes fundamentally play equal roles, enables self-organisation of the CDN and automatic recovery in case of node failures. To optimise the distribution of the content over the different surrogate servers, replica placement algorithms (RPAs) have been developed. In this paper, we present two distributed RPAs for CDNs. We will demonstrate that they further improve CDN performance by reducing the server load and the bandwidth usage. The introduction of link costs allows these algorithms to additionally support load balancing on the network links.

Towards Efficient Service Placement and Server Selection for Large-Scale Deployments

Currently many service providers offer their services using a private and proprietary hard- and software infrastructure. These infrastructures often share many similarities. Hence we believe a generic service delivery architecture, that allows service providers to offer a large array of different services on a shared infrastructure, would provide many advantages over current silo-based approaches. In this paper we propose the first step towards such an architecture, namely several algorithms for dynamically allocating server and network resources to a set of services and selecting a suitable service instance for each client. Service instances are placed on a set of servers, taking into account network resources (available bandwidth), server resources (CPU and memory) and service quality of service (QoS) demands (maximum transmission delay and bandwidth requirements). The optimization goal is to maximize the percentage of satisfied demand (answered requests) and minimize the total number of used servers for service hosting. Each service has a relative priority, which can be manually adjusted to influence the placement scheme.