From d8d6880ef1793f855f5ff6bf5758e260ddaf9258 Mon Sep 17 00:00:00 2001 From: lihangyu <15605149486@163.com> Date: Fri, 19 May 2023 17:29:10 +0800 Subject: [PATCH] [Doc](point query) Modify the document to add constraints and limitations for optimizing high-concurrency point queries. (#19852) --- docs/en/docs/query-acceleration/hight-concurrent-point-query.md | 2 ++ .../docs/query-acceleration/hight-concurrent-point-query.md | 1 + 2 files changed, 3 insertions(+) diff --git a/docs/en/docs/query-acceleration/hight-concurrent-point-query.md b/docs/en/docs/query-acceleration/hight-concurrent-point-query.md index 76a3b98d37..bee73deaab 100644 --- a/docs/en/docs/query-acceleration/hight-concurrent-point-query.md +++ b/docs/en/docs/query-acceleration/hight-concurrent-point-query.md @@ -73,6 +73,8 @@ PROPERTIES ( 1. `enable_unique_key_merge_on_write` should be enabled, since we need primary key for quick point lookup in storage engine 2. when condition only contains primary key like `select * from tbl_point_query where key = 123`, such query will go through the short fast path 3. `light_schema_change` should also been enabled since we rely on `column unique id` of each column when doing a point query. +4. It only supports equality queries on the key column of a single table and does not support joins or nested subqueries. The WHERE condition should consist of the key column alone and be an equality comparison. It can be considered as a type of key-value query. + ## Using `PreparedStatement` diff --git a/docs/zh-CN/docs/query-acceleration/hight-concurrent-point-query.md b/docs/zh-CN/docs/query-acceleration/hight-concurrent-point-query.md index 8f96216604..077cd9d6e3 100644 --- a/docs/zh-CN/docs/query-acceleration/hight-concurrent-point-query.md +++ b/docs/zh-CN/docs/query-acceleration/hight-concurrent-point-query.md @@ -70,6 +70,7 @@ PROPERTIES ( 1. `enable_unique_key_merge_on_write`应该被开启， 存储引擎需要根据主键来快速点查 2. 当条件只包含主键时，如`select * from tbl_point_query where key = 123`，类似的查询会走短路径来优化查询 3. `light_schema_change`应该被开启， 因为主键点查的优化依赖了轻量级 Schema Change 中的`column unique id`来定位列 +4. 只支持单表key列等值查询不支持join、嵌套子查询， **where条件里需要有且仅有key列的等值**， 可以认为是一种key value查询 ## 使用 `PreparedStatement`