mirror of
https://github.com/deepseek-ai/3FS.git
synced 2026-11-05 00:47:47 +00:00
[GH-ISSUE #117] One Idea to Boost DeepSeek 3FS Performance and Simplify Architecture #67
Labels
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
deepseek-ai/3FS#67
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @wpleonardo on GitHub (Mar 6, 2025).
Original GitHub issue: https://github.com/deepseek-ai/3FS/issues/117
I am reaching out on behalf of the OceanBase Community to propose a potential collaboration between OceanBase and DeepSeek 3FS. After reading the article https://mp.weixin.qq.com/s/JbC4YiEj1u1BrBejmiytsA , we believe that OceanBase's capabilities can significantly enhance DeepSeek 3FS in terms of performance, scalability, and architecture optimization.
OceanBase offers features that align well with DeepSeek 3FS’s needs:
OceanBase’s vectorized execution engine can greatly enhance the performance of DeepSeek 3FS, particularly in high-concurrency, compute-heavy tasks like metadata queries, monitoring data analysis, vector search, and checkpointing.
OceanBase’s multi-tenant architecture allows DeepSeek 3FS to consolidate multiple components (e.g., FoundationDB for metadata, ClickHouse for monitoring, and etcd for coordination) into a single OceanBase cluster, managed through separate tenants. This approach can simplify architecture by replacing multiple distributed systems with a unified platform, reducing operational costs through easier deployment and maintenance. It also optimizes resource utilization by dynamically allocating CPU, memory, and storage across tenants based on workload demands.
We’d love to discuss this further and explore the details! Could you please add me as a WeChat friend? We can conduct a technical deep dive and collaborate on performance benchmarking.
My WeChat ID is enjoysunshine0537
Looking forward to your feedback!
@z47z commented on GitHub (Mar 6, 2025):
Wonderful!!!!!!!
@wpleonardo commented on GitHub (Mar 10, 2025):
Hi @z47z and @SF-Zhou , I wanted to follow up on our previous discussion about a potential collaboration between OceanBase and DeepSeek 3FS. I haven’t heard back, so I just wanted to check in—would love to know your thoughts and whether there’s any interest on your side.
If it’s more convenient, feel free to reach out via email (jiuchen.wp@oceanbase.com) or WeChat (WeChat ID: enjoysunshine0537), whichever works best for you. Looking forward to your reply!
@wpleonardo commented on GitHub (Mar 12, 2025):
Hi @wlxiong @yuyuyu101 @level077 , I just wanted to follow up on my previous message regarding a potential collaboration between OceanBase and DeepSeek 3FS. I understand you might be busy, but I’d really appreciate any updates on your thoughts.
If there are any concerns or if now isn’t the right time, feel free to let me know. Also, if email or WeChat is more convenient, you can reach me at email (jiuchen.wp@oceanbase.com) or WeChat (ID: enjoysunshine0537). Looking forward to your response!