[2607.06868] Seekable OCI: Lazy-Loading Container Images via Range-Request Indexing

Skip to main content

arXiv is now an independent nonprofit!<br>Learn more<br>&times;

Search arXiv

Press Enter to search &middot; Advanced search

-->

Computer Science > Distributed, Parallel, and Cluster Computing

arXiv:2607.06868 (cs)

[Submitted on 7 Jul 2026]

Title:Seekable OCI: Lazy-Loading Container Images via Range-Request Indexing

Authors:James Thompson, Wayne Mesard, Jesse Butler, Sri Saran Balaji Rajakumar, Henry Wang<br>View a PDF of the paper titled Seekable OCI: Lazy-Loading Container Images via Range-Request Indexing, by James Thompson and 4 other authors

View PDF<br>HTML (experimental)

Abstract:Container image pulling accounts for the majority of pod startup time in Kubernetes environments. Standard pull downloads the entire image before the container can start, even when the application accesses only a fraction of the image content at startup. We present SOCI (Seekable OCI), a lazy-loading architecture that enables containers to start without downloading the full image. SOCI builds an external index over standard OCI images, mapping files to byte ranges within compressed layers. At runtime, a FUSE filesystem intercepts file accesses and serves them via HTTP range requests. Unlike prior approaches that require image format conversion, SOCI works with unmodified images and standard registries. The index is stored as an OCI referrer artifact, requiring no changes to images, registries, or deployment tooling. On a 1.3 GB Python web service image, SOCI reduces cold-start pull time from 20 seconds to approximately 2.8 seconds (7.4x speedup), with pull time independent of image size. Larger images see larger speedups (9.3x on a 2.5 GB image) because SOCI pull time is constant while standard pull scales linearly. We measure a crossover at 80% access density: below this, lazy loading wins; above, parallel full pull is faster. SOCI lazy loading is deployed in production on Amazon EKS and Amazon ECS Fargate (which launched 18.4 million tasks per day during Prime Day 2025), and has been serving lazy-load requests since 2023. EKS Auto Mode uses SOCI's parallel pull mode for GPU instances.

Subjects:

Distributed, Parallel, and Cluster Computing (cs.DC); Operating Systems (cs.OS)

Cite as:<br>arXiv:2607.06868 [cs.DC]

(or<br>arXiv:2607.06868v1 [cs.DC] for this version)

https://doi.org/10.48550/arXiv.2607.06868

Focus to learn more

arXiv-issued DOI via DataCite (pending registration)

Submission history<br>From: Sri Saran Balaji Vellore Rajakumar [view email]<br>[v1]<br>Tue, 7 Jul 2026 23:55:57 UTC (40 KB)

Full-text links:<br>Access Paper:

View a PDF of the paper titled Seekable OCI: Lazy-Loading Container Images via Range-Request Indexing, by James Thompson and 4 other authors<br>View PDF<br>HTML (experimental)<br>TeX Source

view license

Current browse context:

cs.DC

next >

new<br>recent<br>| 2026-07

Change to browse by:

cs<br>cs.OS

References & Citations

NASA ADS<br>Google Scholar

Semantic Scholar

export BibTeX citation<br>Loading...

BibTeX formatted citation

&times;

loading...

Data provided by:

Bookmark

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Which authors of this paper are endorsers? |<br>Disable MathJax (What is MathJax?)

Major funding support from