morningman opened a new pull request, #4161:
URL: https://github.com/apache/doris-website/pull/4161

   ## What
   
   User documentation for **Peer cache read** in compute-storage decoupled mode 
(apache/doris#63818, picked to branch-4.1 in apache/doris#66167): on a local 
File Cache miss the BE first reads the block from another BE — same or other 
compute group — and only then falls back to object storage; optionally a 
designated "fill" compute group pulls the block from object storage on the 
requester's behalf.
   
   This PR is the **zh-CN 4.x round** for review. The English 4.x page is a 
placeholder so the page routes and can sit in the sidebar; the full English 
translation and the `current` (dev) mirrors follow after review.
   
   ## Changes
   
   - **New page** `compute-storage-decoupled/file-cache/file-cache-peer-read`
     - zh-CN 4.x: scenarios, scope & limits, how it works (candidate sources / 
rotation / cooldown / server fill), enable steps (incl. `SHOW BACKENDS` → 
`compute_group_id`), 4 parameter tables, side effects, Profile counters, BE 
bvars, `/api/peer_cache` HTTP ops, best practices, FAQ
     - EN 4.x: placeholder (definition + version box + figure)
     - Figure 
`static/images/next/compute-storage-decoupled/peer-cache-read-flow.jpg` 
(doris-img-gen style)
   - **Sidebar** `versioned_sidebars/version-4.x-sidebars.json`: added under 
*File Cache* after *File Cache Internals*
   - **Cross-links (zh-CN 4.x)**
     - `compute-storage-decoupled/intro.mdx`: new card
     - `file-cache/file-cache.md`: intro, warm-up section, Profile metrics, 
FAQ, keywords
     - `file-cache/file-cache-internals.md`: manual warm-up contrast, rebalance 
window note, query path, scaling, FAQ
     - `managing-compute-cluster.md`: isolation caution + the previously 
undocumented `peer_read_async_warmup` balance type row
     - `rw/file-cache-rw-compute-group-best-practice.md`: "跨计算组 Peer 读(兜底)" 
subsection + selection note
     - `admin-manual/config/be-config.md`: `enable_cache_read_from_peer` 
rewritten + 17 new peer parameters
     - `query-acceleration/caching-intro.mdx`: new "存算分离文件缓存" card section 
(File Cache + Peer read)
   
   All defaults, metric names and HTTP fields were checked against the 
branch-4.1 source (tag 4.1.4). Version wording is "自 4.2.0 版本起支持" throughout, 
per the doc owner's decision.
   
   ## Not in this PR (next round)
   
   - Full English 4.x page (replaces the placeholder)
   - zh-CN / EN `current` mirrors and the matching `sidebars.ts` entry
   - Open question: `admin-manual/cluster-management/upgrade.md` still lists 
the `enable_cache_read_from_peer` default change under 4.1.4
   
   ## Validation
   
   Static checks only (relative links, anchors and image paths resolve; `git 
diff --check`; 4.x sidebar JSON parses). No site build was run.
   
   🤖 Generated with [Claude Code](https://claude.com/claude-code)
   
   https://claude.ai/code/session_01GRr7pw6ymrf99x3DUr2zxg
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to