提示加速從MySQL數據庫的多個連接

Question

我有一個包含超過1/2百萬行的去年的天氣值（每分鐘1行）的數據庫。 我寫了一系列的SQL查詢，每天查找列的最大值或最小值等，然後找到它發生的時間。然後將數據寫入臨時表，同時找到下一個變量的相同信息。其中一個變量已寫入臨時表，查詢連接表以提供每日高點和低點表。 這需要約4分鐘的時間。有沒有辦法在我的編碼中更高效？儘管我沒有在這方面進行培訓，但這是我所能想到的最好的。 謝謝！ 丹尼爾

CREATE TEMPORARY TABLE IF NOT EXISTS table1 AS (SELECT a.date, a.time AS max_temp_time, a.temperature as max_temp 
FROM Live a 
INNER JOIN (

SELECT DATE, MAX(temperature) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.temperature = aMax.MTotal 
GROUP BY DATE 

             ); 

Drop table if exists records.Daily_Records;            
CREATE TEMPORARY TABLE IF NOT EXISTS table2 AS (SELECT a.date as mindewdate, a.time AS min_dewpoint_time, a.dew_point_temperature as min_dew 
FROM Live a 
INNER JOIN (

SELECT DATE, MIN(dew_point_temperature) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.dew_point_temperature = aMax.MTotal 
GROUP BY mindewdate 

             ); 


CREATE TEMPORARY TABLE IF NOT EXISTS table3 AS (SELECT a.date as maxdewdate, a.time AS max_dewpoint_time, a.dew_point_temperature as max_dew 
FROM Live a 
INNER JOIN (

SELECT DATE, MAX(dew_point_temperature) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.dew_point_temperature = aMax.MTotal 
GROUP BY maxdewdate 

             ); 


CREATE TEMPORARY TABLE IF NOT EXISTS table4 AS (SELECT a.date as mintempdate, a.time AS min_temp_time, a.temperature as min_temp 
FROM Live a 
INNER JOIN (

SELECT DATE, MIN(temperature) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.temperature = aMax.MTotal 
GROUP BY mintempdate 

             ); 



CREATE TEMPORARY TABLE IF NOT EXISTS table5 AS (SELECT a.date as wchilldate, a.time AS windchill_time, a.windchill 
FROM Live a 
INNER JOIN (

SELECT DATE, MIN(windchill) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.windchill = aMax.MTotal 
GROUP BY wchilldate 

             ); 


CREATE TEMPORARY TABLE IF NOT EXISTS table6 AS (SELECT a.date as heat_index_date, a.time AS heat_index_time, a.heat_index 
FROM Live a 
INNER JOIN (

SELECT DATE, MAX(heat_index) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.heat_index = aMax.MTotal 
GROUP BY heat_index_date 

             ); 


CREATE TEMPORARY TABLE IF NOT EXISTS table7 AS (SELECT a.date as max_outdoor_humidity_date, a.time AS max_hum_time, a.outdoor_humidity as max_hum 
FROM Live a 
INNER JOIN (

SELECT DATE, MAX(outdoor_humidity) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.outdoor_humidity = aMax.MTotal 
GROUP BY max_outdoor_humidity_date 

             ); 


CREATE TEMPORARY TABLE IF NOT EXISTS table8 AS (SELECT a.date as  min_outdoor_humidity_date, a.time AS min_hum_time, a.outdoor_humidity as min_hum 
FROM Live a 
INNER JOIN (

SELECT DATE, MIN(outdoor_humidity) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.outdoor_humidity = aMax.MTotal 
GROUP BY min_outdoor_humidity_date 

             ); 

CREATE TEMPORARY TABLE IF NOT EXISTS table9 AS (SELECT a.date as max_rain_rate_date, a.time AS rainrate_time, a.rain_rate as rainrate 
FROM Live a 
INNER JOIN (

SELECT DATE, MAX(rain_rate) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.rain_rate = aMax.MTotal 
GROUP BY max_rain_rate_date 

             ); 


CREATE TEMPORARY TABLE IF NOT EXISTS table10 AS (SELECT a.date as max_daily_rainfall_date, a.time AS max_daily_rainfall_time, a.daily_rainfall as rainfall 
FROM Live a 
INNER JOIN (

SELECT DATE, MAX(daily_rainfall) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.daily_rainfall = aMax.MTotal 
GROUP BY max_daily_rainfall_date 

             ); 


CREATE TEMPORARY TABLE IF NOT EXISTS table11 AS (SELECT a.date as max_gust_windspeed_date, a.time AS gust_time, a.gust_windspeed as gust 
FROM Live a 
INNER JOIN (

SELECT DATE, MAX(gust_windspeed) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.gust_windspeed = aMax.MTotal 
GROUP BY max_gust_windspeed_date 

             ); 


CREATE TEMPORARY TABLE IF NOT EXISTS table12 AS (SELECT a.date as max_average_windspeed_date, a.time AS wind_time, a.average_windspeed as wind 
FROM Live a 
INNER JOIN (

SELECT DATE, MAX(average_windspeed) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.average_windspeed = aMax.MTotal 
GROUP BY max_average_windspeed_date 

             );  

CREATE TEMPORARY TABLE IF NOT EXISTS table13 AS (SELECT a.date as max_barometer_date, a.time AS max_barometer_time, a.barometer as max_barometer 
FROM Live a 
INNER JOIN (

SELECT DATE, MAX(barometer) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.barometer = aMax.MTotal 
GROUP BY max_barometer_date 

             ); 


CREATE TEMPORARY TABLE IF NOT EXISTS table14 AS (SELECT a.date as min_barometer_date, a.time AS min_barometer_time, a.barometer as min_barometer 
FROM Live a 
INNER JOIN (

SELECT DATE, MIN(barometer) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.barometer = aMax.MTotal 
GROUP BY min_barometer_date 

             ); 


CREATE TEMPORARY TABLE IF NOT EXISTS table15 AS (SELECT a.date as max_actual_solar_reading_date, a.time AS max_solar_time, a.actual_solar_reading as solar 
FROM Live a 
INNER JOIN (

SELECT DATE, MAX(actual_solar_reading) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.actual_solar_reading = aMax.MTotal 
GROUP BY max_actual_solar_reading_date 

             );    



CREATE TEMPORARY TABLE IF NOT EXISTS table16 AS (SELECT a.date as davis_vp_uv_date, a.time AS max_uv_time, a.davis_vp_uv as uv 
FROM Live a 
INNER JOIN (

SELECT DATE, MAX(davis_vp_uv) MTotal 
FROM Live 
GROUP BY DATE 
)aMax ON a.date = aMax.date 
AND a.davis_vp_uv = aMax.MTotal 
GROUP BY davis_vp_uv_date 

             ); 




DROP TABLE IF EXISTS records.Daily_Records;            
CREATE TABLE records.Daily_Records AS (SELECT date, max_temp, max_temp_time,min_temp,min_temp_time, windchill,windchill_time, max_dew, max_dewpoint_time, min_dew,min_dewpoint_time,heat_index,heat_index_time,max_hum,max_hum_time,min_hum,min_hum_time,rainrate,rainrate_time,rainfall,max_daily_rainfall_time,gust,gust_time,wind,wind_time,max_barometer,max_barometer_time,min_barometer,min_barometer_time,solar,max_solar_time,uv,max_uv_time 
FROM `table1`          
JOIN `table2` on table2.mindewdate = table1.date 
JOIN `table3` on table3.maxdewdate = table1.date 
JOIN `table4` on table4.mintempdate = table1.date 
JOIN `table5` on table5.wchilldate = table1.date 
JOIN `table6` on table6.heat_index_date = table1.date 
JOIN `table7` on table7.max_outdoor_humidity_date = table1.date 
JOIN `table8` on table8.min_outdoor_humidity_date = table1.date 
JOIN `table10` on table10.max_daily_rainfall_date = table1.date 
JOIN `table11` on table11.max_gust_windspeed_date = table1.date 
JOIN `table12` on table12.max_average_windspeed_date = table1.date 
JOIN `table13` on table13.max_barometer_date = table1.date 
JOIN `table14` on table14.min_barometer_date = table1.date 
JOIN `table15` on table15.max_actual_solar_reading_date = table1.date 
JOIN `table16` on table16.davis_vp_uv_date = table1.date); 

Answer 1

檢查Does MySQL index foreign key columns automatically?

使用InnoDB引擎，所有的外鍵將被適當的支持和索引，導致重要的速度提升。

Answer 2

指數所有的外鍵鏈接表