Technical Report: Bundling Linked Data Structures for Linearizable Range Queries

TL;DR

This paper introduces bundled references, a novel technique enabling linearizable range queries in concurrent linked data structures, significantly improving performance in mixed workloads and database integration.

Contribution

The paper proposes bundled references as a new building block for linearizable range queries, applicable to linked list, skip list, and binary search tree data structures.

Findings

Up to 1.8x faster in skip list range queries

Up to 3.7x faster in binary search tree range queries

Up to 40% performance gain in in-memory database integration

Abstract

We present bundled references, a new building block to provide linearizable range query operations for highly concurrent lock-based linked data structures. Bundled references allow range queries to traverse a path through the data structure that is consistent with the target atomic snapshot. We demonstrate our technique with three data structures: a linked list, skip list, and a binary search tree. Our evaluation reveals that in mixed workloads, our design can improve upon the state-of-the-art techniques by 1.2x-1.8x for a skip list and 1.3x-3.7x for a binary search tree. We also integrate our bundled data structure into the DBx1000 in-memory database, yielding up to 40% gain over the same competitors.