User Manual Privacy Policy Disclaimer Contact us
  Advanced SearchBrowse





Fault Tolerant Gradient Clock Synchronization


Bund,  Johannes
Algorithms and Complexity, MPI for Informatics, Max Planck Society;


Lenzen,  Christoph
Algorithms and Complexity, MPI for Informatics, Max Planck Society;


Rosenbaum,  Will
Algorithms and Complexity, MPI for Informatics, Max Planck Society;

External Ressource
No external resources are shared
Fulltext (public)

(Preprint), 383KB

Supplementary Material (public)
There is no public supplementary material available

Bund, J., Lenzen, C., & Rosenbaum, W. (2019). Fault Tolerant Gradient Clock Synchronization. Retrieved from http://arxiv.org/abs/1902.08042.

Cite as: http://hdl.handle.net/21.11116/0000-0003-0CD6-F
Synchronizing clocks in distributed systems is well-understood, both in terms of fault-tolerance in fully connected systems and the dependence of local and global worst-case skews (i.e., maximum clock difference between neighbors and arbitrary pairs of nodes, respectively) on the diameter of fault-free systems. However, so far nothing non-trivial is known about the local skew that can be achieved in topologies that are not fully connected even under a single Byzantine fault. Put simply, in this work we show that the most powerful known techniques for fault-tolerant and gradient clock synchronization are compatible, in the sense that the best of both worlds can be achieved simultaneously. Concretely, we combine the Lynch-Welch algorithm [Welch1988] for synchronizing a clique of $n$ nodes despite up to $f<n/3$ Byzantine faults with the gradient clock synchronization (GCS) algorithm by Lenzen et al. [Lenzen2010] in order to render the latter resilient to faults. As this is not possible on general graphs, we augment an input graph $\mathcal{G}$ by replacing each node by $3f+1$ fully connected copies, which execute an instance of the Lynch-Welch algorithm. We then interpret these clusters as supernodes executing the GCS algorithm, where for each cluster its correct nodes' Lynch-Welch clocks provide estimates of the logical clock of the supernode in the GCS algorithm. By connecting clusters corresponding to neighbors in $\mathcal{G}$ in a fully bipartite manner, supernodes can inform each other about (estimates of) their logical clock values. This way, we achieve asymptotically optimal local skew, granted that no cluster contains more than $f$ faulty nodes, at factor $O(f)$ and $O(f^2)$ overheads in terms of nodes and edges, respectively. Note that tolerating $f$ faulty neighbors trivially requires degree larger than $f$, so this is asymptotically optimal as well.