SinoTechIntel Academic Portal
Official PDF TranslationEngineering Information Technology and Electronic Engineering

GC bypass: decoupling GC from the flash translation layer to eliminate GC-induced long-tail latency inside SSD

Authors: Shiqiang NIE; Jie NIU; Yingzhao SHAO; Xiaobo LI; Mingming ZHANG; Weiguo WU

DOI: 10.1631/ENG_ITEE_2025_0152Status: Verified Translated Edition
Sponsored AdvertisementAd Placement Area
reCAPTCHA Bot Shield Active

Preparing Secure Academic Download

Verifying human reader & generating high-resolution document...

Verifying Document Integrity15s remaining
← Back to Article
Protected by Google reCAPTCHA v3.PrivacyTerms
Sponsored ContentAdSense In-Feed Ad Slot

Key Findings in This Report

• Proposes GC bypass, a novel architecture that decouples garbage collection from the flash translation layer to reduce long-tail latency in SSDs. • Utilizes dedicated controllers and priority-aware scheduling to ensure user I/O preempts low-priority GC operations, mitigating path conflicts. • Achieves up to 25% reduction in the 99.99th percentile long-tail latency compared to the state-of-the-art Venice scheme. • Maintains the benefits of path diversity while effectively isolating critical I/O from background maintenance tasks.