Fault-tolerant clock synchronization

Abstract

This paper gives two simple efficient distributed algorithms: one for keeping clocks in a network synchronized and one for allowing new processors to join the network with their clocks synchronized.The algorithms tolerate both link and node failures of any type. The algorithm for maintaining synchronization will work for arbitrary networks (rather than just completely connected networks) and tolerates any number of processor or communication link faults as long as the correct processors remain connected by fault-free paths. It thus represents an improvement over other clock synchronization algorithms such as [LM1,LM2,LL1]. Our algorithm for allowing new processors to join requires that more than half the processors be correct, a requirement which is provably necessary.

A Peer-to-Peer Filter-Based Algorithm for Internal Clock Synchronization in Presence of Corrupted Processes

“…[16,17,5,7,10]) where processes can either execute point-to-point or broadcast communication primitives (e.g. [12]). Moreover, these protocols require each process reads the local clock of every other node in the system and, when handling byzantine failures, there is a predefined bound on the number of correct processes in the system in order that the clock synchronization is correct (e.g., [9,17,7].…”

Consul: a communication substrate for fault-tolerant distributed programs

“…One way to measure time intervals by is using a function geUime_elapsed{t : time) that returns the time elapsed since the clock showed time t. This function typically compensates for the changes in the clock value due to synchronization. In another approach [Halp84], the notion of a clock is not bound to specific hardware and a processor may possess any number of clocks. In particular, every instance of clock synchronization logically gives rise to a new version of the clock.…”

Network Time Protocol (version 1) specification and implementation

“…Experimental results on measured times and roundtrip delays in the Internet are discussed in [14], [23] and [31]. Other synchronization protocols are discussed in [7], [17], [20] and [28]. NTP uses techniques evolved from both linear and nonlinear synchronization methodology.…”

“…There are two classes of convergence functions, those involving interactive convergence algorithms and those involving interactive consistency algorithms. Interactive convergence algorithms use statistical clustering techniques such as the fault-tolerant average algorithm of [17], the CNV algorithm of [19], the majority-subset algorithm of [22], the egocentric algorithm of [27] and the algorithms in Section 4 of this document.…”

“…These algorithms use an agreement protocol involving successive rounds of readings, possibly relayed and possibly augmented by digital signatures. Examples include the fireworks algorithm of [17] and the optimum algorithm of [30]. However, these algorithms require large numbers of messages, especially when large numbers of clocks are involved, and are designed to detect faults that have rarely been found in the Internet experience.…”