Concerning the size of logical clocks in distributed systems
Information Processing Letters
An efficient implementation of vector clocks
Information Processing Letters
Time, clocks, and the ordering of events in a distributed system
Communications of the ACM
Visualizing the Performance of Parallel Programs
IEEE Software
A visualization-based environment for top-down debugging of parallel programs
IPPS '95 Proceedings of the 9th International Symposium on Parallel Processing
An Offline Algorithm for Dimension-Bound Analysis
ICPP '99 Proceedings of the 1999 International Conference on Parallel Processing
A framework algorithm for dynamic, centralized dimension-bounded timestamps
CASCON '00 Proceedings of the 2000 conference of the Centre for Advanced Studies on Collaborative research
Hi-index | 0.00 |
The vector-clock size necessary to characterize causality in a distributed computation is bounded by the dimension of the partial order induced by that computation. In theory the dimension can be as large as the number of processes in the computation, but in practice it is much smaller. We present an online algorithm to compute the dimension of a distributed computation. This algorithm requires the computation of the critical pairs of the partial order followed by the creation of extensions of the partial order. This is our next step toward the goal of creating an online vector clock whose size is bounded by the dimension of a distributed computation, not by the number of processes.