“Time, Clocks, and the Ordering of Events in a Distributed System” is a paper by Leslie Lamport published in Communications of the ACM, volume 21, number 7, pages 558 to 565, in July 1978. It is among the most frequently cited papers in all of computer science, and on his own publications page Lamport notes it is his most cited work.
The paper starts from a practical observation: in a distributed system there is no single, perfectly accurate clock, so it is often impossible to say which of two events happened first in real time. Lamport’s answer was to define a partial ordering of events called “happens before,” based purely on which events could have causally influenced which others, and then to show how a system of logical clocks (simple counters exchanged along with messages) can implement an ordering consistent with that relation.
Lamport then extends the idea into a total ordering of events and uses it to solve a concrete coordination problem, building a distributed algorithm for mutual exclusion that grants a shared resource fairly. He observes that the same technique generalizes to implementing any system as a replicated state machine, which became one of the central ideas in distributed computing.
The paper later received the 2000 PODC Influential Paper Award and a 2007 ACM SIGOPS Hall of Fame Award, and its ideas were foundational to Lamport’s 2013 Turing Award. It remains a standard starting point for anyone learning how distributed systems reason about time and order.