Poster abstract: Minimizing flow completion times using adaptive routing over inter-datacenter wide area networks

Inter-datacenter networks connect dozens of geographically dispersed datacenters and carry traffic flows with highly variable sizes and different classes. Adaptive flow routing can improve efficiency and performance by assigning paths to new flows according to network status and flow properties. A popular approach widely used for traffic engineering is based on current bandwidth utilization of links. We propose an alternative that reduces bandwidth usage by up to at least 50% and flow completion times by up to at least 40% across various scheduling policies and flow size distributions.

[1]  Ananth Balashankar,et al.  Software Defined Networking , 2019, 2019 19th International Conference on Sciences and Techniques of Automatic Control and Computer Engineering (STA).

[2]  Keith Kirkpatrick,et al.  Software-defined networking , 2013, CACM.

[3]  Mikkel Thorup,et al.  Optimizing OSPF/IS-IS weights in a changing world , 2002, IEEE J. Sel. Areas Commun..

[4]  Srikanth Kandula,et al.  Calendaring for wide area networks , 2014, SIGCOMM.

[5]  Srikanth Kandula,et al.  DCCast: Efficient Point to Multipoint Transfers Across Datacenters , 2017, HotCloud.

[6]  Srikanth Kandula,et al.  Walking the tightrope: responsive yet stable traffic engineering , 2005, SIGCOMM '05.

[7]  Alex C. Snoeren,et al.  Inside the Social Network's (Datacenter) Network , 2015, Comput. Commun. Rev..

[8]  Srikanth Kandula,et al.  QuickCast: Fast and Efficient Inter-Datacenter Transfers Using Forwarding Tree Cohorts , 2018, IEEE INFOCOM 2018 - IEEE Conference on Computer Communications.