Latency-Aware Forwarding for IRON: Latency Support for Back-Pressure Forwarding

We introduce Latency-Aware Forwarding for IRON, the first comprehensive implementation of a networking solution based on Back-Pressure Forwarding (BPF) with broad support for latency-sensitive traffic. IRON, “Intrinsically Resilient Overlay Network”, has been developed to provide the proven properties of BPF in robustness and throughout optimality while offering higher quality-of-service for latency-constrained traffic. Latency-Aware Forwarding adds support for latency-sensitive traffic while maintaining the throughput optimality for latency-insensitive flows. Latency-Aware Forwarding combines a number of advances to a) forward latency-constrained packets along delay-appropriate paths b) reduce the processing time of these packets at each hop. We detail various techniques developed in support of the two goals above and demonstrate that our approach maintains throughput optimality under the tested conditions. We compare Latency-Aware Forwarding to traditional and workconserving BPF and share results from IRON over a physical network, where we show a 233% increase in goodput in delivery of delay-constrained traffic.

[1]  Eleftheria Athanasopoulou,et al.  Back-Pressure-Based Packet-by-Packet Adaptive Routing in Communication Networks , 2013, IEEE/ACM Transactions on Networking.

[2]  Panganamala Ramana Kumar,et al.  Decentralized throughput maximizing policies for deadline-constrained wireless networks , 2015, 2015 54th IEEE Conference on Decision and Control (CDC).

[3]  O. Rajitha,et al.  Backpressure-Based Packet-By-Packet Adaptive Routing For Traffic Management in Communication Networks , 2014 .

[4]  Atilla Eryilmaz,et al.  Heavy-ball: A new approach to tame delay and convergence in wireless network optimization , 2016, IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications.

[5]  Michael J. Neely Delay-Based Network Utility Maximization , 2010, 2010 Proceedings IEEE INFOCOM.

[6]  Longbo Huang,et al.  Delay reduction via Lagrange multipliers in stochastic network optimization , 2011, IEEE Trans. Autom. Control..

[7]  Bhaskar Krishnamachari,et al.  Routing without routes: the backpressure collection protocol , 2010, IPSN '10.

[8]  Leandros Tassiulas,et al.  Stability properties of constrained queueing systems and scheduling policies for maximum throughput in multihop radio networks , 1990, 29th IEEE Conference on Decision and Control.

[9]  Eytan Modiano,et al.  Fairness and optimal stochastic control for heterogeneous networks , 2005, INFOCOM.

[10]  Eytan Modiano,et al.  Loop-Free Backpressure Routing Using Link-Reversal Algorithms , 2017, IEEE/ACM Transactions on Networking.

[11]  Sucha Supittayapornpong,et al.  Achieving utility-delay-reliability tradeoff in stochastic network optimization with finite buffers , 2015, 2015 IEEE Conference on Computer Communications (INFOCOM).

[12]  Ness B. Shroff,et al.  Delay-Based Back-Pressure Scheduling in Multihop Wireless Networks , 2010, IEEE/ACM Transactions on Networking.

[13]  Alexander L. Stolyar,et al.  A Novel Architecture for Reduction of Delay and Queueing Structure Complexity in the Back-Pressure Algorithm , 2011, IEEE/ACM Transactions on Networking.