The Las-Vegas Processor Identity Problem (How and When to Be Unique)

Abstract

We study the classical problem of assigning unique identifiers to identical concurrent processes. In this paper, we consider the asynchronous shared memory model, and the correctness requirement is that upon termination of the algorithm, the processes must have unique IDs always. Our results include tight characterization of the problem in several respects. We call a protocol solving this task Las Vegas if it has finite expected termination time. Our main positive result is the first Las-Vegas protocol that solves the problem. The protocol terminates in O(logn) expected asychronous rounds, using O(n) shared memory space, where n is the number of participating processes. The new protocol improves on all previous solutions simultaneously in running time (exponentially), probability of termination (to 1), and space requirement. The protocol works under the assumption that the asynchronous schedule is oblivious, i.e., independent of the actual unfolding execution. On the negative side, we show that there is no finite-state Las-Vegas protocol for the problem if the schedule may depend on the history of the shared memory (an adaptive schedule). We also show that any Las-Vegas protocol must know n in advance (which implies that crash faults cannot be tolerated) and that the running time is Ω(logn). For the case of an arbitrary (nonoblivious) adversarial schedule, we present a Las-Vegas protocol that uses O(n) unbounded registers. For the read-modify-write model, we present a constant-space deterministic algorithm.