Sequoia: Treeness of Internet Path Metrics

Sequoia aims to make distributed applications network-aware. That is, enable applications to take advantage of the characteristics of the underlying network such as proximity, bandwidth capacity, and topology. It intends to achieve this through the key concept of prediction trees, a virtual topology of the network, where virtual nodes representing routers connect real end hosts, and carefully computed edge weights model path properties such as latency and bandwidth.

The following are some of the network-aware functionalities that Sequoia seeks to provide:

  • Path Property Prediction: Two hosts in the system can estimate path properties (latency, loss rate, available bandwidth) between them through computation without requiring a direct measurement.
  • Server Selection: An external client outside the system can select the closest or best-provisioned server in the system.
  • Hierarchical Partitioning: Hosts can organize into hierarchical clusters based on their position in the network.

Publications

Demo

demo of geographical partitioning and closest-node discovery in Sequoia for PlanetLab nodes.

Talks