Containerized In-Storage Processing and Computing-Enabled Solid-State Drive Disaggregation

Miryeong Kwon; Donghyun Gouk; Eunjee Na; Jiseon Kim; Junhee Kim; Hyein Woo; Eojin Ryu; Hyunkyu Choi; Jinwoo Baek; Hanyeoreum Bae; Mahmut Kandemir; Myoungsoo Jung

doi:10.1109/MM.2025.3574261

Containerized In-Storage Processing and Computing-Enabled Solid-State Drive Disaggregation

Miryeong Kwon
, Donghyun Gouk
, Eunjee Na
, Jiseon Kim
, Junhee Kim
, Hyein Woo
, Eojin Ryu
, Hyunkyu Choi
, Jinwoo Baek
, Hanyeoreum Bae
, Mahmut Kandemir
, Myoungsoo Jung

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

In-storage processing (ISP) minimizes data transfer for analytics but faces challenges in adaptation and disaggregation. We propose DockerSSD, an ISP model that leverages operating system-level virtualization and lightweight firmware to enable containerized data processing directly on solid-state devices. The key features include Ethernet over Non-volatile Memory Express for network-based ISP management, and virtual firmware for secure, efficient container execution. DockerSSD supports disaggregated storage pools, reducing host overhead and enhancing large-scale services like large language model (LLM) inference. It achieves up to 2 × better performance for input–output-intensive workloads, and 7.9 × improvement in distributed LLM inference.

Original language	English (US)
Pages (from-to)	24-35
Number of pages	12
Journal	IEEE Micro
Volume	45
Issue number	6
DOIs	https://doi.org/10.1109/MM.2025.3574261
State	Published - 2025

All Science Journal Classification (ASJC) codes

Software
Hardware and Architecture
Electrical and Electronic Engineering

Access to Document

10.1109/MM.2025.3574261

Cite this

@article{00b238c0e9d449c9a651eb3c393c2304,

title = "Containerized In-Storage Processing and Computing-Enabled Solid-State Drive Disaggregation",

abstract = "In-storage processing (ISP) minimizes data transfer for analytics but faces challenges in adaptation and disaggregation. We propose DockerSSD, an ISP model that leverages operating system-level virtualization and lightweight firmware to enable containerized data processing directly on solid-state devices. The key features include Ethernet over Non-volatile Memory Express for network-based ISP management, and virtual firmware for secure, efficient container execution. DockerSSD supports disaggregated storage pools, reducing host overhead and enhancing large-scale services like large language model (LLM) inference. It achieves up to 2 × better performance for input–output-intensive workloads, and 7.9 × improvement in distributed LLM inference.",

author = "Miryeong Kwon and Donghyun Gouk and Eunjee Na and Jiseon Kim and Junhee Kim and Hyein Woo and Eojin Ryu and Hyunkyu Choi and Jinwoo Baek and Hanyeoreum Bae and Mahmut Kandemir and Myoungsoo Jung",

note = "Publisher Copyright: {\textcopyright} 1981-2012 IEEE.",

year = "2025",

doi = "10.1109/MM.2025.3574261",

language = "English (US)",

volume = "45",

pages = "24--35",

journal = "IEEE Micro",

issn = "0272-1732",

publisher = "IEEE Computer Society",

number = "6",

}

TY - JOUR

T1 - Containerized In-Storage Processing and Computing-Enabled Solid-State Drive Disaggregation

AU - Kwon, Miryeong

AU - Gouk, Donghyun

AU - Na, Eunjee

AU - Kim, Jiseon

AU - Kim, Junhee

AU - Woo, Hyein

AU - Ryu, Eojin

AU - Choi, Hyunkyu

AU - Baek, Jinwoo

AU - Bae, Hanyeoreum

AU - Kandemir, Mahmut

AU - Jung, Myoungsoo

PY - 2025

Y1 - 2025

N2 - In-storage processing (ISP) minimizes data transfer for analytics but faces challenges in adaptation and disaggregation. We propose DockerSSD, an ISP model that leverages operating system-level virtualization and lightweight firmware to enable containerized data processing directly on solid-state devices. The key features include Ethernet over Non-volatile Memory Express for network-based ISP management, and virtual firmware for secure, efficient container execution. DockerSSD supports disaggregated storage pools, reducing host overhead and enhancing large-scale services like large language model (LLM) inference. It achieves up to 2 × better performance for input–output-intensive workloads, and 7.9 × improvement in distributed LLM inference.

AB - In-storage processing (ISP) minimizes data transfer for analytics but faces challenges in adaptation and disaggregation. We propose DockerSSD, an ISP model that leverages operating system-level virtualization and lightweight firmware to enable containerized data processing directly on solid-state devices. The key features include Ethernet over Non-volatile Memory Express for network-based ISP management, and virtual firmware for secure, efficient container execution. DockerSSD supports disaggregated storage pools, reducing host overhead and enhancing large-scale services like large language model (LLM) inference. It achieves up to 2 × better performance for input–output-intensive workloads, and 7.9 × improvement in distributed LLM inference.

UR - https://www.scopus.com/pages/publications/105007759689

UR - https://www.scopus.com/pages/publications/105007759689#tab=citedBy

U2 - 10.1109/MM.2025.3574261

DO - 10.1109/MM.2025.3574261

M3 - Article

AN - SCOPUS:105007759689

SN - 0272-1732

VL - 45

SP - 24

EP - 35

JO - IEEE Micro

JF - IEEE Micro

IS - 6

ER -

Containerized In-Storage Processing and Computing-Enabled Solid-State Drive Disaggregation

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this