Performance Model for Similarity Caching

TL;DR

This paper introduces a novel approximation method, RND-TTL, for estimating the hit rate of the RND-LRU similarity caching policy, extending classic TTL models to better analyze performance in real-world applications.

Contribution

It develops the RND-TTL approximation, a new method to estimate RND-LRU hit rates by tuning a TTL-based cache model to mimic RND-LRU behavior, including an algorithm for parameter tuning.

Findings

The RND-TTL approximation accurately predicts RND-LRU performance on synthetic traces.

The method is effective on real-world data, demonstrating practical applicability.

The approach provides a systematic way to analyze similarity caching policies.

Abstract

Similarity caching allows requests for an item to be served by a similar item. Applications include recommendation systems, multimedia retrieval, and machine learning. Recently, many similarity caching policies have been proposed, like SIM-LRU and RND-LRU, but the performance analysis of their hit rate is still wanting. In this paper, we show how to extend the popular time-to-live approximation in classic caching to similarity caching. In particular, we propose a method to estimate the hit rate of the similarity caching policy RND-LRU. Our method, the RND-TTL approximation, introduces the RND-TTL cache model and then tunes its parameters in such a way to mimic the behavior of RND-LRU. The parameter tuning involves solving a fixed point system of equations for which we provide an algorithm for numerical resolution and sufficient conditions for its convergence. Our approach for approximating the hit rate of RND-LRU is evaluated on both synthetic and real world traces.