Resource Efficiency and Performance Predictability in A Groupwise, Hardware-Prioritized Cache on NVMe SSDs
IEEE Transactions on Computers · 2026-09 · 75 (9) · 3397-3409
DOI: 10.1109/tc.2026.3711088 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Computers, 2026-09. Verfasst von Miao Cai und 2 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Basic Research Program of Jiangsu, Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China.
Abstract (Originalsprache)
Thanks to notable performance and capacity advantages, NVMe SSDs promise an effective cache tier for alleviating the load pressure of back-end storage servers. Compared with DRAM, an NVMe SSD exhibits up to hundreds of times larger capacity but delivers two orders of magnitude smaller bandwidth per gigabyte. This paper reveals that these hardware characteristics challenge fundamental design goals of caches: (1) storing tons of objects in a large-capacity SSD easily induces severe, persistent fragmentation, resulting in low cache space utilization; (2) intensive front-end requests contend for limited SSD bandwidth, causing unpredictable cache lookup latency. To tackle these challenges, this paper presents Gemini. The core of Gemini is heat-informed, group-based object management with a hardware-assisted I/O prioritization mechanism. Specifically, we introduce a tree-based prediction model for agile object grouping and fast reaction to hotspot shift. It provides SSD-friendly, bulk object eviction with a zero-write, remapping-based migration mechanism. In addition, to ensure request performance, we propose a novel I/O model that mitigates bandwidth congestion and prioritizes their processing at the hardware layer. Evaluation results show that Gemini delivers up to 3.8 ${\boldsymbol{\times}}$ × throughput improvement, 2 ${\boldsymbol{\times}}$ × tail latency reduction, and 2.7 ${\boldsymbol{\times}}$ × higher cache space utilization than state-of-the-art caches for both YCSB workloads and production applications.
Förderung
- National Natural Science Foundation of China
- Basic Research Program of Jiangsu
- Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China
- Overseas Expertise Introduction Center for Discipline Innovation of Food Nutrition and Human Health
- 2024 Yangtze River Delta Science and Technology Innovation Community Joint Research Project
Zitieren
APA
Cai, M., Shen, J., & Ye, B. (2026). Resource Efficiency and Performance Predictability in A Groupwise, Hardware-Prioritized Cache on NVMe SSDs. IEEE Transactions on Computers, 75(9), 3397-3409. https://doi.org/10.1109/tc.2026.3711088
Vancouver
Cai M, Shen J, Ye B. Resource Efficiency and Performance Predictability in A Groupwise, Hardware-Prioritized Cache on NVMe SSDs. IEEE Transactions on Computers. 2026;75(9):3397-3409. doi:10.1109/tc.2026.3711088
BibTeX
@article{Cai2026,
author = {Cai, Miao and Shen, Junru and Ye, Baoliu},
title = {Resource Efficiency and Performance Predictability in A Groupwise, Hardware-Prioritized Cache on NVMe SSDs},
journal = {IEEE Transactions on Computers},
year = {2026},
volume = {75},
number = {9},
pages = {3397--3409},
doi = {10.1109/tc.2026.3711088}
}
EUVIMED