Based on a simple object, an i.i.d. sequence of positive integer-valued random variables {an}n∊ℤ, we introduce and study two random structures and their connections. First, a population dynamics, in which each individual is born at time n and dies at time n + an. This dynamics is that of a D/GI/∞ queue, with arrivals at integer times and service times given by {an}n∊ℤ. Second, the directed random graph Tf on ℤ generated by the random map f(n) = n + an. Assuming only that E [a0] < ∞ and P [a0 = 1] > 0, we show that, in steady state, the population dynamics is regenerative, with one individual alive at each regeneration epoch. We identify a unimodular structure in this dynamics. More precisely, Tf is a unimodular directed tree, in which f(n) is the parent of n. This tree has a unique bi-infinite path. Moreover, Tf splits the integers into two categories: ephemeral integers, with a finite number of descendants of all degrees, and successful integers, with an infinite number. Each regeneration epoch is a successful individual such that all integers less than it are its descendants of some order. Ephemeral, successful, and regeneration integers form stationary and mixing point processes on ℤ.

Based on a simple object, an i.i.d. sequence of positive integervalued random variables, {a n } n∈Z , we introduce and study two random structures and their connections. First, a population dynamics, in which each individual is born at time n and dies at time n + a n . This dynamics is that of a D/GI/∞ queue, with arrivals at integer times and service times given by {a n } n∈Z . Second, the directed random graph T f on Z generated by the random map f (n) = n + a n .Only assuming E[a 0 ] < ∞ and P[a 0 = 1] > 0, we show that, in steady state, the population dynamics is regenerative, with one individual alive at each regenerative epochs. We identify a unimodular structure in this dynamics. More precisely, T f is a unimodular directed tree, in which f (n) is the parent of n. This tree has a unique bi-infinite path. Moreover, T f splits the integers into two categories: ephemeral integers, with a finite number of descendants of all degrees, and successful integers, with an infinite number. Each regenerative epoch is a successful individual such that all integers less than it are its descendants of some order. Ephemeral, successful, and regenerative integers form stationary and mixing point processes on Z.

Many tasks are subject to failure before completion. Two of the most common failure recovery strategies are restart and checkpointing. Under restart, once a failure occurs, it is restarted from the beginning. Under checkpointing, the task is resumed from the preceding checkpoint after the failure. We study asymptotic efficiency of restart for an infinite sequence of tasks, whose sizes form a stationary sequence. We define asymptotic efficiency as the limit of the ratio of the total time to completion in the absence of failures over the total time to completion when failures take place. Whether the asymptotic efficiency is positive or not depends on the comparison of the tail of the distributions of the task size and the random variables governing failures. Our framework allows for variations in the failure rates and dependencies between task sizes. We also study a similar notion of asymptotic efficiency for checkpointing when the task is infinite a.s. and the inter-checkpoint times are i.i.d.. Moreover, in checkpointing, when the failures are exponentially distributed, we prove the existence of an infinite sequence of universal checkpoints, which are always used whenever the system starts from any checkpoint that precedes them.