A Concurrency-Optimal Binary Search Tree

TL;DR

This paper introduces a concurrency-optimal binary search tree that accepts all correct concurrent schedules, outperforming existing BSTs by using a novel locking scheme to maximize concurrency without sacrificing correctness.

Contribution

It presents the first concurrency-optimal BST implementation using a new read-write locking scheme that protects edges, enabling maximum accepted schedules.

Findings

Outperforms state-of-the-art BSTs on common workloads

Uses a novel locking scheme for edges and nodes

Maximizes accepted concurrent schedules

Abstract

The paper presents the first \emph{concurrency-optimal} implementation of a binary search tree (BST). The implementation, based on a standard sequential implementation of an internal tree, ensures that every \emph{schedule} is accepted, i.e., interleaving of steps of the sequential code, unless linearizability is violated. To ensure this property, we use a novel read-write locking scheme that protects tree \emph{edges} in addition to nodes. Our implementation outperforms the state-of-the art BSTs on most basic workloads, which suggests that optimizing the set of accepted schedules of the sequential code can be an adequate design principle for efficient concurrent data structures.