Name: Zero-Extraction Cold Starts: How FUSE-Streaming Slashed ComfyUI Cold Starts by 10x - Fog Dong, BentoML
Start: 2025-06-16T14:50:00+0900
End: 2025-06-16T15:20:00+0900

16-17 June
Learn More and Register to Attend

The Sched app allows you to build your schedule but is not a substitute for your event registration. You must be registered for KubeCon + CloudNativeCon Japan 2025 to participate in the sessions. If you have not registered but would like to join us, please go to the event registration page to purchase a registration.

Please note: This schedule is automatically displayed in Japan Standard Time (UTC+9:00). To see the schedule in your preferred timezone, please select from the drop-down menu to the right, above "Filter by Date." The schedule is subject to change and session seating is available on a first-come, first-served basis.

Monday June 16, 2025 14:50 - 15:20 JST

Level 1 | Pegasus Ballroom A+B1

Cold-start delays for GPU-heavy GenAI apps like ComfyUI aren’t just about speed—they’re architectural failures. While others optimize incremental steps, we eliminate entire phases: no image downloads, no layer extraction, no redundant model copies.

We introduce a radical Kubernetes-native pattern: Direct-to-GPU streaming via FUSE-mounted object storage (S3/GCS), bypassing legacy container workflows. By rearchitecting the snapshotter to support seekable, on-demand FUSE streaming, we enable:

- Instant container boot: Models/CUDA dependencies mount directly from object storage, avoiding registry bottlenecks (40MB/s → 900MB/s throughput)
- Zero-extraction overhead: Layers load incrementally via range-optimized fetches, eliminating Zstd unpack/copy latency
- True cold start elimination: ComfyUI pods activate in 90s (vs. 8+ mins) by co-locating model mounting and inference prep

We’ll dissect a live ComfyUI deployment using 100% OSS primitives to hack container internals in the session.

Speakers

Fog Dong

Senior Software Engineer, BentoML

Fog Dong, a Senior Engineer at BentoML, KubeVela maintainer, CNCF Ambassador, and LFAPAC Evangelist, has a rich background in cloud native and AI infra. Previously instrumental in developing Alibaba's large-scale Serverless workflows and Bytedance's cloud-native CI/CD platform, she... Read More →

Monday June 16, 2025 14:50 - 15:20 JST
Level 1 | Pegasus Ballroom A+B1

AI + ML

Content Experience Level Advanced
Presentation Language English

KubeCon + CloudNativeCon Japan 2025

Fog Dong

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!