Logical time: capturing causality in distributed systems
Raynal, M.
Singhal, M.
Rennes Univ.;
This paper appears in: Computer
Publication Date: Feb 1996
Volume: 29,
Issue: 2
On page(s): 49-56
ISSN: 0018-9162
References Cited: 12
CODEN: CPTRB4
INSPEC Accession Number: 5213840
Digital Object Identifier: 10.1109/2.485846
Current Version Published: 2002-08-06
Abstract
Causality is vital in distributed computations. Distributed
systems can determine causality using logical clocks. Human beings use
the concept of causality to plan, schedule, and execute an enterprise,
or to determine a plan's feasibility. In daily life, we use global time
to deduce causality from loosely synchronized clocks such as wrist
watches and wall clocks. But in distributed computing systems, the rate
of event occurrence is several magnitudes higher, and the
event-execution time several magnitudes smaller. If the physical clocks
in these systems are not synchronized precisely the causality relation
between events cannot be captured accurately. However, distributed
systems have no built-in physical time and can only approximate it. This
article presents a general framework of a system of logical clocks in
distributed systems and discusses three methods: scalar, vector and
matrix, for implementing logical time in these systems
Index
Terms
Available to subscribers and IEEE members.
References
Available to subscribers and IEEE members.
Citing Documents
Available to subscribers and IEEE members.
Cart
