Predicting Available Bandwidth of Internet Path with Ultra Metric Space-Based Approaches

Xing, Changyou; Chen, Ming; Li, Yang

doi:10.1109/glocom.2009.5426036

Cited by 7 publications

(13 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, in real-world network topologies this assumption is often violated [19,11]. Following Xing et al [27], we define an asymmetry coefficient ζ ab , which denotes the bandwidth asymmetry between two nodes a and b as follows:…”

Section: Tree Metrics and Bandwidthmentioning

confidence: 99%

“…Thus, systems have been devised to address this problem [8,10,16]. As bandwidth prediction under asymmetric bandwidth distributions is an even more challenging problem, Xing et al [27] propose PathGuru, which embeds distance matrices in several ultra-metric spaces formed by pre-deployed landmarks. Maintaining incoming and outgoing bandwidth vectors, this system can predict asymmetric bandwidth distributions.…”

Section: Related Workmentioning

confidence: 99%

“…The model proposed by Beaumont et al [2] is also able to cope with asymmetry, but assumes the access link to determine the bottleneck bandwidth of each path. Xing et al [27] propose embedding bandwidth in a set of ultrametric spaces, but their system is based on landmarks, and thus, not fully decentralizable.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Bandwidth Prediction in the Face of Asymmetry

Schober

Brenner

Kapitza

et al. 2013

Distributed Applications and Interoperable Systems

View full text Add to dashboard Cite

Abstract. An increasing number of networked applications, like video conference and video-on-demand, benefit from knowledge about Internet path measures like available bandwidth. Server selection and placement of infrastructure nodes based on accurate information about network conditions help to improve the quality-of-service of these systems. Acquiring this knowledge usually requires fully-meshed ad-hoc measurements. These, however, introduce a large overhead and a possible delay in communication establishment. Thus, prediction-based approaches like Sequoia have been proposed, which treat path properties as a semimetric and embed them onto trees, leveraging labelling schemes to predict distances between hosts not measured before. In this paper, we identify asymmetry as a cause of serious distortion in these systems causing inaccurate prediction. We study the impact of asymmetric network conditions on the accuracy of existing tree-embedding approaches, and present direction-aware embedding, a novel scheme that separates upstream from downstream properties of hosts and significantly improves the prediction accuracy for highly asymmetric datasets. This is achieved by embedding nodes for each direction separately and constraining the distance calculation to inversely labelled nodes. We evaluate the effectiveness and trade-offs of our approach using synthetic as well as real-world datasets.

show abstract

Section: Tree Metrics and Bandwidthmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Bandwidth Prediction in the Face of Asymmetry

Schober

Brenner

Kapitza

et al. 2013

Distributed Applications and Interoperable Systems

View full text Add to dashboard Cite

show abstract

“…For each network link between new node to each target node, l i in a candidate metric tree, p i 6 Calculate relative error, e i = |actual measurementpredicted latency|/actual measurement 7 Calculate average relative error, e p 8 9 Identify optimal metric tree, p opt in which p i achieves min…”

Section: Algorithm 4: Constructoptimalmetrictree (T N M)mentioning

confidence: 99%

“…al. [6] presented another bandwidth prediction method by measuring to selected landmark nodes, and then using those measurements to predict to any node. Given node A and B have measurements to all landmark nodes, we can predict the bandwidth from A to B.…”

Section: Related Workmentioning

confidence: 99%

Network latency prediction using high accuracy prediction tree

Ping

Chong

Karuppiah

2013

Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication

View full text Add to dashboard Cite

Network latency is often used as an optimization parameter for network path construction over the Internet for various real-time applications. This paper proposes a high accuracy prediction tree method for latency estimation minimizing the need for intrusive mesh measurements. The network overlay of communication nodes is represented as a tree structure, called a prediction tree, with the latency of unmeasured network links predicted based on selected measured network links. We describe three novel heuristics that are the foundations of this high accuracy prediction tree, assisted by optimal target node selection and elimination of imprecise prediction steps. We have examined the proposed method based on publicly available data to ensure accuracy of high precision latency estimation process. Experiment results show that with 50% measurement, our proposed algorithm obtains 82% accuracy of latency prediction over a 120-node network.

show abstract

NPA-BT: A Network Performance Aware BitTorrent Traffic Optimization Mechanism

Xing

Chen

2010

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. More than half of today's Internet traffic is generated by P2P applications, and most of which is the BitTorrent file sharing traffic. The random neighbor selection mechanism of BT puts unprecedented pressure on the network operators and service providers. In this paper firstly we give a theoretical analysis on the impact of neighbor selection mechanism on BitTorrent performance as well as network resource cost. And then based on network performance prediction techniques, we propose a network performance aware BitTorrent traffic optimization mechanism NPA-BT, which uses network performance information to help overlay constructing. Simulation results show that NPA-BT can select peers in the same AS network and with good performance as neighbors, and thus the BitTorrent download time and the cross AS network traffic are all decreased.

show abstract

Predicting Available Bandwidth of Internet Path with Ultra Metric Space-Based Approaches

Cited by 7 publications

References 10 publications

Bandwidth Prediction in the Face of Asymmetry

Bandwidth Prediction in the Face of Asymmetry

Network latency prediction using high accuracy prediction tree

NPA-BT: A Network Performance Aware BitTorrent Traffic Optimization Mechanism

Contact Info

Product

Resources

About