查詢postgresql耗時最慢的SQL
PostgreSQL部署上之後，經過一段時間的執行，我們比較關心那些SQL執行時間比較長，或者說那些SQL執行的特別慢，拖累的效能，只有找到這些SQL，才能有針對性地對這些SQL進行優化，提升PostgreSQL的效能。
PostgreSQL提供了pg_stat_statements來儲存SQL的執行次數，總執行時間，shared_buffer命中次數，shared_buffer read次數等統計資訊。
| Name | Type | References | Description |
|---|---|---|---|
| userid | oid | pg_authid.oid | OID of user who executed the statement |
| dbid | oid | pg_database.oid | OID of database in which the statement was executed |
| query | text | Text of the statement (up to track_activity_query_size |
|
| calls | bigint | Number of times executed | |
| total_time | double precision | Total time spent in the statement, in seconds | |
| rows | bigint | Total number of rows retrieved or affected by the statement | |
| shared_blks_hit | bigint | Total number of shared blocks hits by the statement | |
| shared_blks_read | bigint | Total number of shared blocks reads by the statement | |
| shared_blks_written | bigint | Total number of shared blocks writes by the statement | |
| local_blks_hit | bigint | Total number of local blocks hits by the statement | |
| local_blks_read | bigint | Total number of local blocks reads by the statement | |
| local_blks_written | bigint | Total number of local blocks writes by the statement | |
| temp_blks_read | bigint | Total number of temp blocks reads by the statement | |
| temp_blks_written | bigint | Total number of temp blocks writes by the statemen |
上圖表來自PostgreSQL官方文件，注意的一點是，我的PostgreSQL是9.1.9,此時total_time的單位是秒，我觀9.2的PostgreSQL的文件，total_time單位已經是毫秒。所以我的參考文獻More On PostgreSQL perform 裡面應該用的是9.2,因為外國這位大神預設單位是毫秒。
可以看出，pg_stat_statements統計了SQL的很多資訊，方便我們分析SQL的效能。但是這個屬於PostgreSQL的擴充套件，需要修改postgresql.conf，才能使用：
操作步驟如下
1 修改配置檔案，並且重啟PostgreSQL方能生效
- #------------------------------------------------------------------------------
- # PG_STAT_STATEMENTS OPTIONS
- #------------------------------------------------------------------------------
- shared_preload_libraries = 'pg_stat_statements'
- custom_variable_classes = 'pg_stat_statements'
- pg_stat_statements.max = 1000
- pg_stat_statements.track = all
2 建立pg_stat_statements擴充套件
- CREATE EXTENSION pg_stat_statements;
從此之後，PostgreSQL就能記錄SQL的統計資訊。
上面的表格雖然豐富，其實我們基本比較關心執行最慢的SQL，如何檢視執行最慢的10條SQL？
SELECT query, calls, total_time, (total_time/calls) as average ,rows,
100.0 * shared_blks_hit /nullif(shared_blks_hit + shared_blks_read, 0) AS hit_percent
FROM pg_stat_statements
ORDER BY average DESC LIMIT 10; 我在我本地的DB，查詢最慢的2條SQL，輸出如下：
在我另一臺機器上，用pgadmin檢視：
統計結果一直都在，重啟也不會清零，那麼統計結果如何清零重新統計呢？執行下面SQL即可：
- select pg_stat_statements_reset() ;
找到最耗時的SQL，我們就能針對這些耗時的SQL，檢視是否有優化的餘地。
參考文獻：
1 More on Postgres Performance
2 PostgreSQL manual