Measurement-based construction of locality-aware overlay networks
Qian Zhang, Wenwu Zhu, Zhensheng Zhang, Xinyan Zhang, Gang Song, Lixin Gao
Abstract
Qian Zhang, Wenwu Zhu, Zhensheng Zhang, Xinyan Zhang, Gang Song, Lixin Gao
Abstract
One important aspect of constructing an overlay network is how to exploit network locality in the underlying network. In this paper, we propose a scalable protocol for constructing an overlay network that takes account of locality of network hosts. The constructed overlay network can significantly decrease the communication cost between end-hosts. Our simulation results show that the average distance between a pair of hosts in the constructed overlay network is only about 11% of the one in a traditional, randomly connected overlay network. Furthermore, our proposed overlay considered to be more scalable than tree-based or mesh-based overlays.
OpenAlex reports 9 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
One important aspect of constructing an overlay network is how to exploit network locality in the underlying network. In this paper, we propose a scalable protocol for constructing an overlay network that takes account of locality of network hosts. The constructed overlay network can significantly decrease the communication cost between end-hosts. Our simulation results show that the average distance between a pair of hosts in the constructed overlay network is only about 11% of the one in a traditional, randomly connected overlay network. Furthermore, our proposed overlay considered to be more scalable than tree-based or mesh-based overlays.
Key concepts: Overlay network, Computer science, Overlay, Locality, Computer network, Scalability, Distributed computing, Exploit