Fast Concurrent Cuckoo Kick-Out Eviction Schemes for High-Density Tables

TL;DR

This paper introduces novel eviction algorithms and concurrency schemes for high-density cuckoo hash tables, significantly improving insertion efficiency and reducing transaction aborts in multi-writer environments.

Contribution

It presents new kick-out eviction algorithms and a claim flag synchronization mechanism, enhancing performance and concurrency in high-density cuckoo hashing.

Findings

Reduced bins viewed per insertion by up to ten times

Decreased transaction aborts in multi-writer scenarios

Improved efficiency for delete-light workloads

Abstract

Cuckoo hashing guarantees constant-time lookups regardless of table density, making it a viable candidate for high-density tables. Cuckoo hashing insertions perform poorly at high table densities, however. In this paper, we mitigate this problem through the introduction of novel kick-out eviction algorithms. Experimentally, our algorithms reduce the number of bins viewed per insertion for high-density tables by as much as a factor of ten. We also introduce an optimistic concurrency scheme for transactional multi-writer cuckoo hash tables (not using hardware transactional memory). For delete-light workloads, one of our kick-out algorithms avoids all competition between insertions with high probability, and significantly reduces transaction-abort frequency. This result is extended to arbitrary workloads using a new synchronization mechanism called a claim flag.