From e7b7d0765d1b37aa0d9d3d5b2fcefca9f6b54fc9 Mon Sep 17 00:00:00 2001 From: Kolbe Kegel Date: Wed, 8 Sep 2021 14:11:59 -0700 Subject: [PATCH] performance_schema -> information_schema This page mistakenly claimed that statements_summary was in performance_schema rather than information_schema. I also added additional clarification about the difference. --- statement-summary-tables.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/statement-summary-tables.md b/statement-summary-tables.md index 368e52e9ba0bc..0037eff67ccbd 100644 --- a/statement-summary-tables.md +++ b/statement-summary-tables.md @@ -7,7 +7,7 @@ summary: Learn about Statement Summary Table in TiDB. To better handle SQL performance issues, MySQL has provided [statement summary tables](https://dev.mysql.com/doc/refman/5.6/en/statement-summary-tables.html) in `performance_schema` to monitor SQL with statistics. Among these tables, `events_statements_summary_by_digest` is very useful in locating SQL problems with its abundant fields such as latency, execution times, rows scanned, and full table scans. -Therefore, starting from v4.0.0-rc.1, TiDB provides system tables in `information_schema`. These system tables are similar to `events_statements_summary_by_digest` in terms of features. +Therefore, starting from v4.0.0-rc.1, TiDB provides system tables in `information_schema` (_not_ `performance_schema`) that are similar to `events_statements_summary_by_digest` in terms of features. - [`statements_summary`](#statements_summary) - [`statements_summary_history`](#statements_summary_history) @@ -19,7 +19,7 @@ This document details these tables and introduces how to use them to troubleshoo ## `statements_summary` -`statements_summary` is a system table in `performance_schema`. `statements_summary` groups the SQL statements by the SQL digest and the plan digest, and provides statistics for each SQL category. +`statements_summary` is a system table in `information_schema`. `statements_summary` groups the SQL statements by the SQL digest and the plan digest, and provides statistics for each SQL category. The "SQL digest" here means the same as used in slow logs, which is a unique identifier calculated through normalized SQL statements. The normalization process ignores constant, blank characters, and is case insensitive. Therefore, statements with consistent syntaxes have the same digest. For example: