kvcache-ai/Mooncake

Fork 0

mirror of https://github.com/kvcache-ai/Mooncake.git synced 2026-11-05 01:17:55 +00:00

[GH-ISSUE #4339] CNCF AI Technical Community Group Discussion/Presentation #1089

New issue

Open

opened 2026-10-29 21:22:45 +00:00 by bsakaguchi · 7 comments

bsakaguchi commented

2026-10-29 21:22:45 +00:00

Owner

Originally created by @julioc-p on GitHub (Sep 26, 2026).
Original GitHub issue: https://github.com/kvcache-ai/Mooncake/issues/4339

Originally assigned to: @stmatengss on GitHub.

Hello Mooncake maintainers,

I'm one of the leads of the CNCF AI Technical Community Group, and we'd love to have you present Mooncake at one of our sessions. We've been seeing it show up in RL setups lately too, on top of the KV cache work.

The audience is mostly people running AI workloads on Kubernetes. Recent sessions covered llm-d, KubeRay and Milvus. Around 30 min plus Q&A, recorded.

We have an APAC-friendly slot on Tuesdays, early afternoon Beijing time, and late October or November would work on our side. Feel free to reply here or find us in #cncf-artificial-intelligence-technical-community-group on CNCF Slack.

Thanks!

cc: @joshhalley

Originally created by @julioc-p on GitHub (Sep 26, 2026). Original GitHub issue: https://github.com/kvcache-ai/Mooncake/issues/4339 Originally assigned to: @stmatengss on GitHub. Hello Mooncake maintainers, I'm one of the leads of the [CNCF AI Technical Community Group](https://community.cncf.io/artificial-intelligence-technical-community-group/), and we'd love to have you present Mooncake at one of our sessions. We've been seeing it show up in RL setups lately too, on top of the KV cache work. The audience is mostly people running AI workloads on Kubernetes. Recent sessions covered llm-d, KubeRay and Milvus. Around 30 min plus Q&A, recorded. We have an APAC-friendly slot on Tuesdays, early afternoon Beijing time, and late October or November would work on our side. Feel free to reply here or find us in [#cncf-artificial-intelligence-technical-community-group](https://slack.cncf.io/) on CNCF Slack. Thanks! cc: @joshhalley

bsakaguchi commented

2026-10-29 21:22:46 +00:00

Author

Owner

@github-actions[bot] commented on GitHub (Sep 26, 2026):

Thanks for opening this issue, @julioc-p!

Field	Value
Issue	#4339
GitHub user ID	`32308064`
Reporter	@julioc-p

A maintainer will triage this when possible. To help us respond faster, please include:

Mooncake version or commit SHA
Environment (OS, CUDA/driver, RDMA stack if relevant)
Steps to reproduce and expected vs. actual behavior

Useful links: Documentation · Contributing guide

This message was posted automatically by the issue bot.

@github-actions[bot] commented on GitHub (Sep 26, 2026): Thanks for opening this issue, @julioc-p! | Field | Value | |-------|-------| | **Issue** | #4339 | | **GitHub user ID** | `32308064` | | **Reporter** | @julioc-p | A maintainer will triage this when possible. To help us respond faster, please include: - Mooncake version or commit SHA - Environment (OS, CUDA/driver, RDMA stack if relevant) - Steps to reproduce and expected vs. actual behavior Useful links: [Documentation](https://kvcache-ai.github.io/Mooncake/) · [Contributing guide](https://github.com/kvcache-ai/Mooncake/blob/main/CONTRIBUTING.md) > This message was posted automatically by the issue bot.

bsakaguchi commented

2026-10-29 21:22:46 +00:00

Author

Owner

@stmatengss commented on GitHub (Sep 28, 2026):

Hello Mooncake maintainers,

I'm one of the leads of the CNCF AI Technical Community Group, and we'd love to have you present Mooncake at one of our sessions. We've been seeing it show up in RL setups lately too, on top of the KV cache work.

The audience is mostly people running AI workloads on Kubernetes. Recent sessions covered llm-d, KubeRay and Milvus. Around 30 min plus Q&A, recorded.

We have an APAC-friendly slot on Tuesdays, early afternoon Beijing time, and late October or November would work on our side. Feel free to reply here or find us in #cncf-artificial-intelligence-technical-community-group on CNCF Slack.

Thanks!

cc: @joshhalley

Hi @julioc-p, thanks for the invitation! We’d be happy to present Mooncake to the CNCF AI community.

@stmatengss commented on GitHub (Sep 28, 2026): > Hello Mooncake maintainers, > > I'm one of the leads of the [CNCF AI Technical Community Group](https://community.cncf.io/artificial-intelligence-technical-community-group/), and we'd love to have you present Mooncake at one of our sessions. We've been seeing it show up in RL setups lately too, on top of the KV cache work. > > The audience is mostly people running AI workloads on Kubernetes. Recent sessions covered llm-d, KubeRay and Milvus. Around 30 min plus Q&A, recorded. > > We have an APAC-friendly slot on Tuesdays, early afternoon Beijing time, and late October or November would work on our side. Feel free to reply here or find us in [#cncf-artificial-intelligence-technical-community-group](https://slack.cncf.io/) on CNCF Slack. > > Thanks! > > cc: [@joshhalley](https://github.com/joshhalley) Hi @julioc-p, thanks for the invitation! We’d be happy to present Mooncake to the CNCF AI community.

bsakaguchi commented

2026-10-29 21:22:46 +00:00

Author

Owner

@julioc-p commented on GitHub (Sep 28, 2026):

Great to hear, thanks! Topics sound great. Candidate dates: Tue Oct 27, Nov 10 or Nov 24, all at 14:00 Beijing time (07:00 CET). Once you pick one and confirm a speaker, I'll set up the event.

@julioc-p commented on GitHub (Sep 28, 2026): Great to hear, thanks! Topics sound great. Candidate dates: Tue Oct 27, Nov 10 or Nov 24, all at 14:00 Beijing time (07:00 CET). Once you pick one and confirm a speaker, I'll set up the event.

bsakaguchi commented

2026-10-29 21:22:46 +00:00

Author

Owner

@stmatengss commented on GitHub (Oct 9, 2026):

3. Great to hear, thanks! Topics sound great. Candidate dates: Tue Oct 27, Nov 10 or Nov 24, all at 14:00 Beijing time (07:00 CET). Once you pick one and confirm a speaker, I'll set up the event.

@julioc-p Nov 10 and Nov 24 look good to me. I will present recent progress in RL and heterogeneous weight updates.

@stmatengss commented on GitHub (Oct 9, 2026): > 3\. Great to hear, thanks! Topics sound great. Candidate dates: Tue Oct 27, Nov 10 or Nov 24, all at 14:00 Beijing time (07:00 CET). Once you pick one and confirm a speaker, I'll set up the event. @julioc-p Nov 10 and Nov 24 look good to me. I will present recent progress in RL and heterogeneous weight updates.

bsakaguchi commented

2026-10-29 21:22:46 +00:00

Author

Owner

@stmatengss commented on GitHub (Oct 11, 2026):

Hi @julioc-p, we'd like to move ahead with the Mooncake session and get involved in the AI TCG's ongoing technical discussions and collaboration.

Following up on my availability above, both November 10 and November 24 at 14:00 Beijing time work for me. Please let me know which date fits your schedule so we can finalize the event.

Speaker: Teng Ma (@stmatengss), Mooncake maintainer.

Proposed title: Mooncake: Shared KV Cache and Data Movement for Distributed AI Inference and RL

Proposed abstract:
Distributed AI workloads need to move and reuse large amounts of data across inference and training workers. This session introduces Mooncake Store and Transfer Engine, covering shared KV cache storage, prefill/decode disaggregation, and high-performance data transfer. We will discuss integration with vLLM and SGLang, Kubernetes deployment through RBG, and recent progress in RL rollout data movement and heterogeneous weight updates. The discussion will focus on architecture, practical deployment considerations, and opportunities for collaboration across the cloud-native AI ecosystem.

Proposed outline (30 minutes, followed by Q&A):

5 minutes: data movement and cache reuse in distributed AI workloads.
10 minutes: Mooncake Store, Transfer Engine, and inference integrations.
10 minutes: recent RL and heterogeneous weight-update progress.
5 minutes: Kubernetes deployment considerations and collaboration opportunities.

Please also point me to the current recurring meeting calendar and any active initiatives around distributed inference, shared KV caching, or RL data movement where Mooncake contributors could help.

Project and documentation: https://github.com/kvcache-ai/Mooncake · https://kvcache-ai.github.io/Mooncake/

@stmatengss commented on GitHub (Oct 11, 2026): Hi @julioc-p, we'd like to move ahead with the Mooncake session and get involved in the AI TCG's ongoing technical discussions and collaboration. Following up on my availability above, both **November 10 and November 24 at 14:00 Beijing time** work for me. Please let me know which date fits your schedule so we can finalize the event. **Speaker:** Teng Ma (@stmatengss), Mooncake maintainer. **Proposed title:** Mooncake: Shared KV Cache and Data Movement for Distributed AI Inference and RL **Proposed abstract:** Distributed AI workloads need to move and reuse large amounts of data across inference and training workers. This session introduces Mooncake Store and Transfer Engine, covering shared KV cache storage, prefill/decode disaggregation, and high-performance data transfer. We will discuss integration with vLLM and SGLang, Kubernetes deployment through RBG, and recent progress in RL rollout data movement and heterogeneous weight updates. The discussion will focus on architecture, practical deployment considerations, and opportunities for collaboration across the cloud-native AI ecosystem. **Proposed outline (30 minutes, followed by Q&A):** - 5 minutes: data movement and cache reuse in distributed AI workloads. - 10 minutes: Mooncake Store, Transfer Engine, and inference integrations. - 10 minutes: recent RL and heterogeneous weight-update progress. - 5 minutes: Kubernetes deployment considerations and collaboration opportunities. Please also point me to the current recurring meeting calendar and any active initiatives around distributed inference, shared KV caching, or RL data movement where Mooncake contributors could help. Project and documentation: https://github.com/kvcache-ai/Mooncake · https://kvcache-ai.github.io/Mooncake/

bsakaguchi commented

2026-10-29 21:22:46 +00:00

Author

Owner

@julioc-p commented on GitHub (Oct 11, 2026):

Thanks @stmatengss, title and abstract look great! Let's go with Nov 24 at 14:00 Beijing time, I'll set up the event and share the link here. Our sessions are listed at https://ocgroups.dev/cncf/group/6auhr3u, discussions happen in #cncf-artificial-intelligence-technical-community-group on CNCF Slack, and the CNCF AI initiatives are tracked here: https://github.com/cncf/toc/issues?q=state%3Aopen+label%3A%22toc%2Finitiative%2FAI%22

@julioc-p commented on GitHub (Oct 11, 2026): Thanks @stmatengss, title and abstract look great! Let's go with Nov 24 at 14:00 Beijing time, I'll set up the event and share the link here. Our sessions are listed at https://ocgroups.dev/cncf/group/6auhr3u, discussions happen in #cncf-artificial-intelligence-technical-community-group on CNCF Slack, and the CNCF AI initiatives are tracked here: https://github.com/cncf/toc/issues?q=state%3Aopen+label%3A%22toc%2Finitiative%2FAI%22

bsakaguchi commented

2026-10-29 21:22:46 +00:00

Author

Owner

@julioc-p commented on GitHub (Oct 14, 2026):

Here's the event: https://ocgroups.dev/cncf/group/6auhr3u/event/e5xxdrh. Register there and you'll get the join link by email. See you on Nov 24!

@julioc-p commented on GitHub (Oct 14, 2026): Here's the event: https://ocgroups.dev/cncf/group/6auhr3u/event/e5xxdrh. Register there and you'll get the join link by email. See you on Nov 24!

bsakaguchi referenced this issue

2026-10-29 21:32:04 +00:00

[PR #1089] [MERGED] [CI] add sglang integration test #2005

bsakaguchi referenced this issue

2026-10-29 21:32:04 +00:00

[PR #1090] [CLOSED] [WIP] Move test-sglang-integration job to standalone workflow file #2008

bsakaguchi referenced this issue

2026-10-29 21:32:26 +00:00

[PR #1180] [CLOSED] [CI] add sglang e2e integration test #2057