Performance and Cost-Aware Cache Provisioning
2026-08-10 • Performance
Performance
AI summaryⓘ
The authors studied how to best use cache storage in edge-cloud systems by focusing on meeting specific performance goals rather than just fixing cache size. They looked at the smallest cache needed to achieve a desired hit rate and also considered the processing time involved. They found that adjusting parts of the caching method based on past usage helps save space and reduce work. Their new hybrid policy performs better by needing less storage while keeping computation low, tested on real and simulated data.
cache policycache hit rateedge-cloudservice-level objective (SLO)cache capacitysegmented policiesworkload patternsprocessing costhybrid cache policy
Authors
Ridwanul Tanvir, George Kesidis
Abstract
While traditional cache policy evaluations fix capacity - often at 0.1% of the dataset - and measure the resulting hit rate, practical edge-cloud deployments require balancing both storage and computational overhead as billed resources. Consequently, system operators frequently focus on a different objective: determining the minimum cache size needed to satisfy a specific Service-Level Objective (SLO) hit-rate. This paper explores this SLO-centric paradigm by analyzing the minimum capacity and execution time each policy requires to hit a defined target. Additionally, we show that dynamically adjusting the segment ratio in segmented policies based on historical workload patterns enhances efficiency. Through evaluations across real-world and synthetic traces, we present a novel hybrid segmented policy that reduces capacity requirements while keeping processing costs low.