Abstract
SUMMARYWe present a new software-based clock synchronization scheme that provides high precision time agreement among distributed memory nodes. The technique is designed to minimize variance from a reference chimer during runtime and with minimal time-request latency. Our scheme permits initial unbounded variations in time and corrects both slow and fast chimers (clock skew). An implementation developed within the context of the message passing interface is described, and time coordination measurements are presented. Among our results, the mean time variance for a set of nodes improved from 20.0 ms under standard Network Time Protocol down to 2.29 μs under our scheme.
Original language | English |
---|---|
Pages (from-to) | 881-897 |
Number of pages | 17 |
Journal | Concurrency and Computation: Practice and Experience |
Volume | 25 |
Issue number | 6 |
DOIs | |
State | Published - Apr 25 2013 |
Keywords
- MPI
- clock synchronization
- programming tools
- supercomputing
- system software
- time service