Paxos Made Wireless: Consensus in the Air

Many applications in low-power wireless networks require complex coordination between their members. Swarms of robots or sensors and actuators in industrial closed-loop control need to coordinate within short periods of time to execute tasks. Failing to agree on a common decision can cause substantial consequences, like system failures and threats to human life. Such applications require consensus algorithms to enable coordination. While consensus has been studied for wired networks decades ago, with, for example, Paxos and Raft, it remains an open problem in multi-hop low-power wireless networks due to the limited resources available and the high cost of established solutions. This paper presents Wireless Paxos, a fault-tolerant, network-wide consensus primitive for low-power wireless networks. It is a new flavor of Paxos, the most-used consensus protocol today, and is specifically designed to tackle the challenges of low-power wireless networks. By building on top of concurrent transmissions, it provides low-latency, high reliability, and guarantees on the consensus. Our results show that Wireless Paxos requires only 289 ms to complete a consensus between 188 nodes in testbed experiments. Furthermore, we show that Wireless Paxos stays consistent even when injecting controlled failures.

[1]  Leslie Lamport,et al.  Fast Paxos , 2006, Distributed Computing.

[2]  Jim Gray,et al.  Notes on Data Base Operating Systems , 1978, Advanced Course: Operating Systems.

[3]  Leslie Lamport,et al.  Paxos Made Simple , 2001 .

[4]  Robert Griesemer,et al.  Paxos made live: an engineering perspective , 2007, PODC '07.

[5]  James Brown,et al.  Constructing Schedules for Time-Critical Data Delivery in Wireless Sensor Networks , 2014, TOSN.

[6]  Fernando Pedone,et al.  Ring Paxos: A high-throughput atomic broadcast protocol , 2010, 2010 IEEE/IFIP International Conference on Dependable Systems & Networks (DSN).

[7]  Michael Isard,et al.  Autopilot: automatic data center management , 2007, OPSR.

[8]  Mun Choon Chan,et al.  Splash : Fast Data Dissemination with Constructive Interference in Wireless Sensor Networks , 2013 .

[9]  Lothar Thiele,et al.  Virtual Synchrony Guarantees for Cyber-physical Systems , 2013, 2013 IEEE 32nd International Symposium on Reliable Distributed Systems.

[10]  Jun Rao,et al.  Using Paxos to Build a Scalable, Consistent, and Highly Available Datastore , 2011, Proc. VLDB Endow..

[11]  Federico Ferrari,et al.  Chaos: versatile and efficient all-to-all data sharing and in-network processing at scale , 2013, SenSys '13.

[12]  Adam Dunkels,et al.  Contiki - a lightweight and flexible operating system for tiny networked sensors , 2004, 29th Annual IEEE International Conference on Local Computer Networks.

[13]  K. Leentvaar,et al.  The Capture Effect in FM Receivers , 1976, IEEE Trans. Commun..

[14]  Christopher Frost,et al.  Spanner: Google's Globally-Distributed Database , 2012, OSDI.

[15]  Pascale Launay,et al.  Solving Consensus in Opportunistic Networks , 2015, ICDCN.

[16]  Murat Demirbas,et al.  Consensus and collision detectors in wireless Ad Hoc networks , 2005, PODC '05.

[17]  John K. Ousterhout,et al.  In Search of an Understandable Consensus Algorithm , 2014, USENIX ATC.

[18]  Nancy A. Lynch,et al.  Impossibility of distributed consensus with one faulty process , 1983, PODS '83.

[19]  Adam Dunkels,et al.  The ContikiMAC Radio Duty Cycling Protocol , 2011 .

[20]  Lothar Thiele,et al.  Low-power wireless bus , 2012, SenSys '12.

[21]  Andreas Terzis,et al.  Design and evaluation of a versatile and efficient receiver-initiated link layer for low-power wireless , 2010, SenSys '10.

[22]  Jean-Philippe Martin,et al.  Fast Byzantine Consensus , 2006, IEEE Transactions on Dependable and Secure Computing.

[23]  Miguel Oom Temudo de Castro,et al.  Practical Byzantine fault tolerance , 1999, OSDI '99.

[24]  Amy L. Murphy,et al.  Data Prediction + Synchronous Transmissions = Ultra-low Power Wireless Sensor Networks , 2016, SenSys.

[25]  Federico Ferrari,et al.  FlockLab: A testbed for distributed, synchronized tracing and profiling of wireless embedded systems , 2013, 2013 ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN).

[26]  Michael Stonebraker,et al.  A Formal Model of Crash Recovery in a Distributed System , 1983, IEEE Transactions on Software Engineering.

[27]  Marco Zuniga,et al.  JAG: Reliable and Predictable Wireless Agreement under External Radio Interference , 2012, 2012 IEEE 33rd Real-Time Systems Symposium.

[28]  Andreas Köpke.  Engineering a communication protocol stack to support consensus in sensor networks , 2013 .

[29]  Philip Levis,et al.  The β-factor: measuring wireless link burstiness , 2008, SenSys '08.

[30]  Eric Fleury,et al.  FIT IoT-LAB: A large scale open experimental IoT testbed , 2015, 2015 IEEE 2nd World Forum on Internet of Things (WF-IoT).

[31]  Leslie Lamport,et al.  The part-time parliament , 1998, TOCS.

[32]  David E. Culler,et al.  Versatile low power media access for wireless sensor networks , 2004, SenSys '04.

[33]  Leslie Lamport,et al.  Cheap Paxos , 2004, International Conference on Dependable Systems and Networks, 2004.

[34]  Aníbal Ollero,et al.  Journal of Intelligent & Robotic Systems manuscript No. (will be inserted by the editor) Experimental Results in Multi-UAV Coordination for Disaster Management and Civil Security Applications , 2022 .

[35]  Olaf Landsiedel,et al.  Network-wide Consensus Utilizing the Capture Effect in Low-power Wireless Networks , 2017, SenSys.

[36]  André Schiper,et al.  Extending Paxos/LastVoting with an Adequate Communication Layer for Wireless Ad Hoc Networks , 2008, 2008 Symposium on Reliable Distributed Systems.

[37]  Miguel Correia,et al.  Byzantine Fault-Tolerant Consensus in Wireless Ad Hoc Networks , 2013, IEEE Transactions on Mobile Computing.

[38]  Eric Anderson,et al.  X-MAC: a short preamble MAC protocol for duty-cycled wireless sensor networks , 2006, SenSys '06.

[39]  Lothar Thiele,et al.  Efficient network flooding and time synchronization with Glossy , 2011, Proceedings of the 10th ACM/IEEE International Conference on Information Processing in Sensor Networks.