Skip to content
Projects
Groups
Snippets
Help
This project
Loading...
Sign in / Register
Toggle navigation
P
pac
Overview
Overview
Details
Activity
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Charts
Wiki
Wiki
Snippets
Snippets
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
ai-tools
pac
Commits
96efdf58
Commit
96efdf58
authored
Aug 30, 2026
by
luoqi
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
docs(gap): 测试机验收协议 —— 交替四轮避免缓存偏袒
parent
f15b183a
Pipeline
#3619
failed in 0 seconds
Changes
1
Pipelines
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
19 additions
and
0 deletions
+19
-0
docs/design/gap-set-based-rewrite-plan.md
+19
-0
No files found.
docs/design/gap-set-based-rewrite-plan.md
View file @
96efdf58
...
...
@@ -269,3 +269,22 @@ WHERE cardinality(COALESCE(rem.remaining_teeth, ARRAY[]::text[])) > 0
### 对拍工具自检
`--self`
(legacy 对 legacy)先跑过,零差异 —— 证明工具本身不会把相同当不同。
## 11. 测试机验收协议(2026-08-30 夜)
脚本
`~/gapverify.sh`
(测试机),产物落
`~/gapbench/`
。
| 步 | 内容 | 门 |
|---|---|---|
| 0 | 工具自检:
`--self --sub=jaw_cyst`
(legacy 对 legacy) | 必须零差异,否则后面的"零差异"不可信 |
| 1 | 画像消费方:
`--persona=2000`
| 零差异 + 单患者 p95 不劣化 >10% |
| 2 | 召回 11 子场景:逐 (患者×信号×牙位) 双向 EXCEPT ALL |
**零差异**
(不是"差异很少") |
| 3 | 交替全量重算 A(legacy) → B(setbased) → C(legacy) → D(setbased) | 行级 diff 全 0;取 C vs D(都是热态第二轮)比墙钟 |
**为什么要交替四轮**
:第一轮把页读进缓存,第二轮才代表稳态。只跑 legacy→setbased
会让 setbased 白蹭前一轮的缓存,把收益量虚高 —— 这条分支上已经有两次"理论漂亮实测为零"
的前科,不能再拿有偏的数字下结论。
**噪音源**
:测试机
`PAC_STALE_SCAN_CRON=0 2 * * *`
(画像 stale 扫描)会落在验收中段。
C/D 相邻,长时间后台负载对两者影响相近,不单边偏袒;若数字明显抖动,重跑 C/D 一对。
`PAC_INCREMENTAL_CRON=15 8 * * *`
在窗口之外。
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment