SlimIO: Lightweight I/O Path Design for Write Isolation in FDP-backed In-Memory Databases

Sangyun Lee; Sungjin Byeon; Soon Hwang; Jaewan Park; Joo Young Hwang; Junyoung Han; Javier González; Awais Khan; Youngjae Kim

doi:10.1145/3731599.3767511

SlimIO: Lightweight I/O Path Design for Write Isolation in FDP-backed In-Memory Databases

Sangyun Lee
, Sungjin Byeon
, Soon Hwang
, Jaewan Park
, Joo Young Hwang
, Junyoung Han
, Javier González
, Awais Khan
, Youngjae Kim

Technology Integration

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In-Memory Databases (IMDBs) are widely used with HPC applications to manage transient data, often using snapshot-based persistence for backups. Redis, a representative IMDB, employs both snapshot and Write-Ahead Log (WAL) mechanisms, storing data on persistent devices via the traditional kernel I/O path. This method incurs syscall overhead, I/O contention between processes, and SSD garbage collection (GC) delays. To address these issues, we propose SlimIO, which adopts I/O passthru to minimize syscall overhead and inter-process I/O interference. Additionally, it leverages Flexible Data Placement (FDP) SSDs as backup storage to avoid performance degradation from SSD GC. Experimental results show that SlimIO reduces snapshot time by up to 25%, increases query throughput by up to 30% during non-snapshot periods, and lowers 99.9%-ile latency by up to 50%. Furthermore, it achieves a write amplification factor (WAF) of 1.00, indicating no redundant internal writes, thus extending SSD lifespan.

Original language	English
Title of host publication	Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops
Publisher	Association for Computing Machinery, Inc
Pages	1375-1384
Number of pages	10
ISBN (Electronic)	9798400718717
DOIs	https://doi.org/10.1145/3731599.3767511
State	Published - Nov 15 2025
Event	2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops - St. Louis, United States Duration: Nov 16 2025 → Nov 21 2025

Publication series

Name	Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops

Conference

Conference	2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops
Country/Territory	United States
City	St. Louis
Period	11/16/25 → 11/21/25

Funding

This work was partially supported by Samsung Electronics Co., Ltd. (IO221014-02908-01) and the National Research Foundation of Korea (NRF), funded by the Korean government (MSIT), under Grant No. RS-2025-00564249. This research also used resources of the Oak Ridge Leadership Computing Facility, located at the National Center for Computational Sciences at the Oak Ridge National Laboratory, which is supported by the Office of Science of the DOE under Contract DE-AC05-00OR22725.

Keywords

FDP SSDs
In-memory database
Snapshot

Access to Document

10.1145/3731599.3767511

Cite this

Lee, S., Byeon, S., Hwang, S., Park, J., Hwang, J. Y., Han, J., González, J., Khan, A., & Kim, Y. (2025). SlimIO: Lightweight I/O Path Design for Write Isolation in FDP-backed In-Memory Databases. In Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops (pp. 1375-1384). (Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops). Association for Computing Machinery, Inc. https://doi.org/10.1145/3731599.3767511

Lee, Sangyun ; Byeon, Sungjin ; Hwang, Soon et al. / SlimIO : Lightweight I/O Path Design for Write Isolation in FDP-backed In-Memory Databases. Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops. Association for Computing Machinery, Inc, 2025. pp. 1375-1384 (Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops).

@inproceedings{1c5ea5ab214b40e78865acb74d0c1a72,

title = "SlimIO: Lightweight I/O Path Design for Write Isolation in FDP-backed In-Memory Databases",

abstract = "In-Memory Databases (IMDBs) are widely used with HPC applications to manage transient data, often using snapshot-based persistence for backups. Redis, a representative IMDB, employs both snapshot and Write-Ahead Log (WAL) mechanisms, storing data on persistent devices via the traditional kernel I/O path. This method incurs syscall overhead, I/O contention between processes, and SSD garbage collection (GC) delays. To address these issues, we propose SlimIO, which adopts I/O passthru to minimize syscall overhead and inter-process I/O interference. Additionally, it leverages Flexible Data Placement (FDP) SSDs as backup storage to avoid performance degradation from SSD GC. Experimental results show that SlimIO reduces snapshot time by up to 25\%, increases query throughput by up to 30\% during non-snapshot periods, and lowers 99.9\%-ile latency by up to 50\%. Furthermore, it achieves a write amplification factor (WAF) of 1.00, indicating no redundant internal writes, thus extending SSD lifespan.",

keywords = "FDP SSDs, In-memory database, Snapshot",

author = "Sangyun Lee and Sungjin Byeon and Soon Hwang and Jaewan Park and Hwang, \{Joo Young\} and Junyoung Han and Javier Gonz{\'a}lez and Awais Khan and Youngjae Kim",

note = "Publisher Copyright: {\textcopyright} 2025 Copyright held by the owner/author(s).; 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops ; Conference date: 16-11-2025 Through 21-11-2025",

year = "2025",

month = nov,

day = "15",

doi = "10.1145/3731599.3767511",

language = "English",

series = "Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops",

publisher = "Association for Computing Machinery, Inc",

pages = "1375--1384",

booktitle = "Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops",

}

Lee, S, Byeon, S, Hwang, S, Park, J, Hwang, JY, Han, J, González, J, Khan, A & Kim, Y 2025, SlimIO: Lightweight I/O Path Design for Write Isolation in FDP-backed In-Memory Databases. in Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops. Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops, Association for Computing Machinery, Inc, pp. 1375-1384, 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops, St. Louis, United States, 11/16/25. https://doi.org/10.1145/3731599.3767511

SlimIO: Lightweight I/O Path Design for Write Isolation in FDP-backed In-Memory Databases. / Lee, Sangyun; Byeon, Sungjin; Hwang, Soon et al.
Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops. Association for Computing Machinery, Inc, 2025. p. 1375-1384 (Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - SlimIO

T2 - 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops

AU - Lee, Sangyun

AU - Byeon, Sungjin

AU - Hwang, Soon

AU - Park, Jaewan

AU - Hwang, Joo Young

AU - Han, Junyoung

AU - González, Javier

AU - Khan, Awais

AU - Kim, Youngjae

PY - 2025/11/15

Y1 - 2025/11/15

N2 - In-Memory Databases (IMDBs) are widely used with HPC applications to manage transient data, often using snapshot-based persistence for backups. Redis, a representative IMDB, employs both snapshot and Write-Ahead Log (WAL) mechanisms, storing data on persistent devices via the traditional kernel I/O path. This method incurs syscall overhead, I/O contention between processes, and SSD garbage collection (GC) delays. To address these issues, we propose SlimIO, which adopts I/O passthru to minimize syscall overhead and inter-process I/O interference. Additionally, it leverages Flexible Data Placement (FDP) SSDs as backup storage to avoid performance degradation from SSD GC. Experimental results show that SlimIO reduces snapshot time by up to 25%, increases query throughput by up to 30% during non-snapshot periods, and lowers 99.9%-ile latency by up to 50%. Furthermore, it achieves a write amplification factor (WAF) of 1.00, indicating no redundant internal writes, thus extending SSD lifespan.

AB - In-Memory Databases (IMDBs) are widely used with HPC applications to manage transient data, often using snapshot-based persistence for backups. Redis, a representative IMDB, employs both snapshot and Write-Ahead Log (WAL) mechanisms, storing data on persistent devices via the traditional kernel I/O path. This method incurs syscall overhead, I/O contention between processes, and SSD garbage collection (GC) delays. To address these issues, we propose SlimIO, which adopts I/O passthru to minimize syscall overhead and inter-process I/O interference. Additionally, it leverages Flexible Data Placement (FDP) SSDs as backup storage to avoid performance degradation from SSD GC. Experimental results show that SlimIO reduces snapshot time by up to 25%, increases query throughput by up to 30% during non-snapshot periods, and lowers 99.9%-ile latency by up to 50%. Furthermore, it achieves a write amplification factor (WAF) of 1.00, indicating no redundant internal writes, thus extending SSD lifespan.

KW - FDP SSDs

KW - In-memory database

KW - Snapshot

UR - https://www.scopus.com/pages/publications/105023408955

U2 - 10.1145/3731599.3767511

DO - 10.1145/3731599.3767511

M3 - Conference contribution

AN - SCOPUS:105023408955

T3 - Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops

SP - 1375

EP - 1384

BT - Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops

PB - Association for Computing Machinery, Inc

Y2 - 16 November 2025 through 21 November 2025

ER -

Lee S, Byeon S, Hwang S, Park J, Hwang JY, Han J et al. SlimIO: Lightweight I/O Path Design for Write Isolation in FDP-backed In-Memory Databases. In Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops. Association for Computing Machinery, Inc. 2025. p. 1375-1384. (Proceedings of 2025 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC 2025 Workshops). doi: 10.1145/3731599.3767511

SlimIO: Lightweight I/O Path Design for Write Isolation in FDP-backed In-Memory Databases

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this