Tiny-tail flash: Near-perfect elimination of garbage collection tail latencies in NAND SSDs

S Yan, H Li, M Hao, MH Tong… - ACM Transactions on …, 2017 - dl.acm.org
Flash storage has become the mainstream destination for storage users. However, SSDs do
not always deliver the performance that users expect. The core culprit of flash performance …

The {CASE} of {FEMU}: Cheap, accurate, scalable and extensible flash emulator

H Li, M Hao, MH Tong, S Sundararaman… - … USENIX Conference on …, 2018 - usenix.org
We present FEMU, a QEMU-based flash emulator for fostering future full-stack
software/hardware SSD research, with the following four" CASE" benefits. FEMU is cheap …

{LinnOS}: Predictability on unpredictable flash storage with a light neural network

M Hao, L Toksoz, N Li, EE Halim, H Hoffmann… - … USENIX Symposium on …, 2020 - usenix.org
This paper presents LinnOS, an operating system that leverages a light neural network for
inferring SSD performance at a very fine—per-IO—granularity and helps parallel storage …

{FlashBlox}: Achieving Both Performance Isolation and Uniform Lifetime for Virtualized {SSDs}

J Huang, A Badam, L Caulfield, S Nath… - … USENIX Conference on …, 2017 - usenix.org
A longstanding goal of SSD virtualization has been to provide performance isolation
between multiple tenants sharing the device. Virtualizing SSDs, however, has traditionally …

The tail at store: A revelation from millions of hours of disk and {SSD} deployments

M Hao, G Soundararajan… - … USENIX Conference on …, 2016 - usenix.org
We study storage performance in over 450,000 disks and 4,000 SSDs over 87 days for an
overall total of 857 million (disk) and 7 million (SSD) drive hours. We find that storage …

IODA: A host/device co-design for strong predictability contract on modern flash storage

H Li, ML Putra, R Shi, X Lin, GR Ganger… - Proceedings of the ACM …, 2021 - dl.acm.org
Predictable latency on flash storage is a long-pursuit goal, yet, unpredictability stays due to
the unavoidable disturbance from many well-known SSD internal activities. To combat this …

Alleviating garbage collection interference through spatial separation in all flash arrays

J Kim, K Lim, Y Jung, S Lee, C Min… - 2019 USENIX Annual …, 2019 - usenix.org
We present SWAN, a novel All Flash Array (AFA) management scheme. Recent flash SSDs
provide high I/O bandwidth (eg, 3-10GB/s) so the storage bandwidth can easily surpass the …

Access characteristic guided read and write cost regulation for performance improvement on flash memory

Q Li, L Shi, CJ Xue, K Wu, C Ji, Q Zhuge… - … USENIX Conference on …, 2016 - usenix.org
The relatively high cost of write operations has become the performance bottleneck of flash
memory. Write cost refers to the time needed to program a flash page using incremental-step …

RackBlox: A Software-Defined Rack-Scale Storage System with Network-Storage Co-Design

B Reidys, Y Xue, D Li, B Sukhwani, WM Hwu… - Proceedings of the 29th …, 2023 - dl.acm.org
Software-defined networking (SDN) and software-defined flash (SDF) have been serving as
the backbone of modern data centers. They are managed separately to handle I/O requests …

{FusionRAID}: Achieving Consistent Low Latency for Commodity {SSD} Arrays

T Jiang, G Zhang, Z Huang, X Ma, J Wei, Z Li… - … USENIX Conference on …, 2021 - usenix.org
The use of all-flash arrays has been increasing. Compared to their hard-disk counterparts,
each drive offers higher performance but also undergoes more severe periodic performance …