River wubx

Databend · cloud

[River wubx @wubx]

@DashHuang 原来老罗和我一样，也是直接关注你了。

[Evot.AI @Evot_AI]

We just shipped 𝙀𝙫𝙤𝙩.𝘼𝙄, and open-sourced the runtime behind it.

It's a 𝙙𝙞𝙨𝙩𝙧𝙞𝙗𝙪𝙩𝙚𝙙 𝘼𝙄 𝙩𝙚𝙖𝙢 𝙥𝙡𝙖𝙩𝙛𝙤𝙧𝙢 built for teams and enterprises. Multiple agents running in parallel, sharing the same context and memory, getting smarter together over time.

Here's the problem we kept running into. You give one agent a complex task, it gets slow, it loses context halfway, and it starts from scratch every time. The obvious fix is more agents. But then they're all islands -- Agent A figured out your deployment process last week, and Agent B has no idea.

So we built a 𝙨𝙝𝙖𝙧𝙚𝙙 𝙙𝙖𝙩𝙖 𝙡𝙖𝙮𝙚𝙧 underneath. Every agent runs on its own node with isolated compute, but they all read and write to the same memory, the same knowledge base, the same history. When one agent learns something, it's there for everyone on the next run. A brand new agent on day one already has everything the team has ever figured out.

What makes it different:
- 𝙎𝙝𝙖𝙧𝙚𝙙 𝙗𝙧𝙖𝙞𝙣 -- all agents connected to one data layer. Context, memory, knowledge, history, not siloed.
- 𝗗𝗶𝘀𝘁𝗿𝗶𝗯𝘂𝘁𝗲𝗱 𝗯𝘆 𝗱𝗲𝗳𝗮𝘂𝗹𝘁 -- agents dispatch subtasks across nodes. One overloaded? It hands off to a peer.
- 𝙇𝙤𝙘𝙖𝙡 + 𝘾𝙡𝙤𝙪𝙙 -- local nodes touch your files and dev environment, cloud nodes handle heavy compute. One cluster, no boundaries.
- 𝙎𝙚𝙡𝙛-𝙚𝙫𝙤𝙡𝙫𝙞𝙣𝙜 -- every completed run deposits knowledge back into the shared pool. No prompt tuning, no manual handoff.
- 𝙁𝙪𝙡𝙡 𝙩𝙧𝙖𝙘𝙚 𝙖𝙣𝙙 𝙖𝙪𝙙𝙞𝙩 -- every operation recorded end to end. Who did what, which tools fired, what came back. All queryable.
- 𝙋𝙧𝙤𝙙𝙪𝙘𝙩𝙞𝙤𝙣-𝙧𝙚𝙖𝙙𝙮 -- secrets vault-isolated, approval gates on sensitive actions, per-agent token budgets with proactive alerts.
- 100+ 𝙞𝙣𝙩𝙚𝙜𝙧𝙖𝙩𝙞𝙤𝙣𝙨 -- Slack, Lark, GitHub, Linear, and counting. Internal tools? Write a Skill, point it at a script. Done.

Humans set direction, review what matters, approve what's critical. 𝙀𝙫𝙚𝙧𝙮𝙩𝙝𝙞𝙣𝙜 𝙞𝙣 𝙗𝙚𝙩𝙬𝙚𝙚𝙣 𝙞𝙨 𝙩𝙝𝙚 𝘼𝙄 𝙩𝙚𝙖𝙢'𝙨 𝙟𝙤𝙗.

The engine underneath is 𝘽𝙚𝙣𝙙𝘾𝙡𝙖𝙬 -- distributed AgentOS, written in Rust, open source as of today.

Come try it: https://t.co/6NECxlFzOs
Star the repo: https://t.co/M1Ta2UPwhH

[River wubx @wubx]

看了一下 OpenAI 在数据库上优化实践 https://t.co/yA56ah81NT 感觉是传统 DBA 的手活全被他们实践了一下。

[River wubx @wubx]

@DatabendLabs It also support Stream, another type of table cdc in Databend.

[Databend @DatabendLabs]

Meet Fuse Engine—our powerful storage engine built from scratch in 2021, inspired by Apache Iceberg. Purely S3-based, no disk or caching needed, yet excels at:
✅ Petabyte-scale analytics with 10,000+ QPS
✅ Sub-second real-time transactions (single & multi-table)
✅ Zero-copy schema evolution for instant table changes
✅ Built-in Time Travel for data versioning & audits
✅ Automatic vacuum, compaction, reclustering—no manual maintenance!

[River wubx @wubx]

This one very rude meetup. Rust VS C.
https://t.co/XQZKZCRuVT

[River wubx @wubx]

@zhaiwx1987 Oracle officials wondered why not support the MySQL protocol in Oracle？
What the hell do you want？

[NanoVMs @nanovms]

this week on the unikernel application spotlight we explore running @DatabendLabs - the open source snowflake alternative as nanos unikernels - https://t.co/JAHdc9brZL - check it out!

[River wubx @wubx]

SlateDB is an embedded storage engine built as a log-structured merge-tree. Unlike traditional LSM-tree storage engines, SlateDB writes data to object storage (S3, GCS, ABS, MinIO, Tigris, and so on). #NextDatabase

[River wubx @wubx]

#SlateDB is an embedded storage engine built as a log-structured merge-tree. Unlike traditional LSM-tree storage engines, SlateDB writes data to object storage (S3, GCS, ABS, MinIO, Tigris, and so on). #NextDatabase

https://t.co/8VJeNd1HN8

[Databend @DatabendLabs]

🚀 #Databend is now available in DBeaver 24.3.1! 🎉
Huge thanks to rad-pat from the community. 🙌
You can now find it as a new connection type under:
𝙉𝙚𝙬 𝘿𝙖𝙩𝙖𝙗𝙖𝙨𝙚 𝘾𝙤𝙣𝙣𝙚𝙘𝙩𝙞𝙤𝙣 → 𝘼𝙣𝙖𝙡𝙮𝙩𝙞𝙘𝙖𝙡 → 𝘿𝙖𝙩𝙖𝙗𝙚𝙣𝙙
Check out the release: https://t.co/r5FV368TiP
@dbeaver_news

[Databend @DatabendLabs]

Thanks to @github for the support! #Databend Labs' GitHub org is now databendlabs (formerly datafuselabs): https://t.co/toiCoIpqNo. Databend has reached 7.7k stars and serves as an open-source alternative to #Snowflake, written in #Rust.

[Databend @DatabendLabs]

Customer Success Story: How Typing Achieved 90% Cost Savings in Big Data with Databend Cloud https://t.co/AuNEQySqsR

[sundyli @sundyli1]

Inspired by the blog of #duckdb , #databend implemented a new aggregate hash table to handle group aggregation.
Furthermore, we added several optimizations tailored for distributed scenarios.

https://t.co/QH661lV1Iq

[Xuanwo @OnlyXuanwo]

😍, Nice work from @sundyli1 to improve the @DatabendLabs 's hyperloglog and ndv implementations that inspired by @Redisinc and @otmar_ertl's paper: "New cardinality estimation algorithms for HyperLogLog sketches"

https://t.co/aKNc1T1oLb

[River wubx @wubx]

@MarkCallaghanDB @lefred @PeterZaitsev No, Cloud native must have:
1. Pay-as-you-go
2. Storage separation
3. Native High availability

[River wubx @wubx]

@BohuTANG Kind of like dblink

[River wubx @wubx]

@MarkCallaghanDB This scenario works well for the Databend Cloud https://t.co/Ihf3ZUctAZ. @BuhuTang

[sundyli @sundyli1]

Brilliant idea of adaptive filter reordering in query engine

https://t.co/O9mVUgtCii

[River wubx @wubx]

Databend lateral flatten is very friendly to json presentation.
https://t.co/yokuCivkfC