An Empirical Comparison of FAISS and FENSHSES for Nearest Neighbor Search in Hamming Space

TL;DR

This paper empirically compares FAISS and FENSHSES for nearest neighbor search in Hamming space, analyzing their performance trade-offs in indexing speed, search latency, and memory usage to inform system design choices.

Contribution

It provides a comprehensive empirical evaluation of FAISS and FENSHSES, highlighting their performance differences and trade-offs in main memory versus secondary memory implementations.

Findings

FAISS has faster indexing speed than FENSHSES.

FENSHSES consumes less RAM during search.

Trade-offs identified between speed and memory usage.

Abstract

In this paper, we compare the performances of FAISS and FENSHSES on nearest neighbor search in Hamming space--a fundamental task with ubiquitous applications in nowadays eCommerce. Comprehensive evaluations are made in terms of indexing speed, search latency and RAM consumption. This comparison is conducted towards a better understanding on trade-offs between nearest neighbor search systems implemented in main memory and the ones implemented in secondary memory, which is largely unaddressed in literature.