EXPLAIN指令

EXPLAIN指令是檢視優化器如何決定執行查詢的主要方法，但最終不一定就是按照這個來。

EXPLAIN可以作用于 SELECT、DELETE、INSERT、UPDATE和REPLACE語句。

可以使用FORMAT=JSON來輸出詳細的執行計劃成本

大緻是以下的格式

{
    "query_block": {
        "select_id": 1,
        "cost_info": {
            "query_cost": "399.75"
        },
        "table": {
            "table_name": "zc_order",
            "access_type": "index",
            "key": "IDX_ORDER_NO",
            "used_key_parts": [
                "ORDER_NO"
            ],
            "key_length": "98",
            "rows_examined_per_scan": 3755,
            "rows_produced_per_join": 3755,
            "filtered": "100.00",
            "using_index": true,
            "cost_info": {
                "read_cost": "24.25",
                "eval_cost": "375.50",
                "prefix_cost": "399.75",
                "data_read_per_join": "14M"
            },
            "used_columns": [
                "ID"
            ]
        }
    }
}

執行計劃包含的資訊

+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+------------------------------+
| id | select_type | table | partitions | type | possible_keys | key  | key_len | ref  | rows | filtered | Extra                        |
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+------------------------------+

詳見：MySQL官方文檔-explain-output

下面具體看看每一項的含義

id

包含一組數字，表示查詢中執行select子句或操作表的順序

id相同，表示同一組，執行順序由上至下

id不同則值越大優先級越高，越先執行

如果是子查詢，id的序号會遞增，id值越大優先級越高，越先被執行

臨時表的id是NULL

select_type

查詢中每個select子句的類型（簡單OR複雜）

select_type對于非SELECT語句，顯示的是其類型，例如UPDATE顯示UPDATE

SIMPLE：簡單的SELECT，查詢中不包含子查詢或者UNION

mysql> EXPLAIN SELECT * FROM zc_order;
+----+-------------+----------+------------+------+---------------+------+---------+------+------+----------+-------+
| id | select_type | table    | partitions | type | possible_keys | key  | key_len | ref  | rows | filtered | Extra |
+----+-------------+----------+------------+------+---------------+------+---------+------+------+----------+-------+
|  1 | SIMPLE      | zc_order | NULL       | ALL  | NULL          | NULL | NULL    | NULL | 3781 |      100 | NULL  |
+----+-------------+----------+------------+------+---------------+------+---------+------+------+----------+-------+

PRIMARY：最外層的SELECT，查詢中若包含任何複雜的子部分，最外層查詢則被标記

SUBQUERY：在SELECT或WHERE清單中包含了子查詢，該子查詢被标記

mysql> EXPLAIN SELECT * from zc_order_goods where order_no = (select MAX(order_no) from zc_order);
+----+-------------+----------------+------------+------+---------------+------+---------+------+------+----------+------------------------------+
| id | select_type | table          | partitions | type | possible_keys | key  | key_len | ref  | rows | filtered | Extra                        |
+----+-------------+----------------+------------+------+---------------+------+---------+------+------+----------+------------------------------+
|  1 | PRIMARY     | zc_order_goods | NULL       | ALL  | NULL          | NULL | NULL    | NULL | 4696 |       10 | Using where                  |
|  2 | SUBQUERY    | NULL           | NULL       | NULL | NULL          | NULL | NULL    | NULL | NULL | NULL     | Select tables optimized away |
+----+-------------+----------------+------------+------+---------------+------+---------+------+------+----------+------------------------------+

UNION：标記出現在UNION後的SELECT

UNION RESULT：UNION的結果

mysql> EXPLAIN SELECT * from zc_order o where o.order_no = 'UNO200418000000008' UNION SELECT * from zc_order o2 where o2.order_no = 'UNO200418000000009';
+------+--------------+------------+------------+-------+---------------+--------------+---------+-------+------+----------+-----------------+
| id   | select_type  | table      | partitions | type  | possible_keys | key          | key_len | ref   | rows | filtered | Extra           |
+------+--------------+------------+------------+-------+---------------+--------------+---------+-------+------+----------+-----------------+
|    1 | PRIMARY      | o          | NULL       | const | IDX_ORDER_NO  | IDX_ORDER_NO | 130     | const |    1 |      100 | NULL            |
|    2 | UNION        | o2         | NULL       | const | IDX_ORDER_NO  | IDX_ORDER_NO | 130     | const |    1 |      100 | NULL            |
| NULL | UNION RESULT | <union1,2> | NULL       | ALL   | NULL          | NULL         | NULL    | NULL  | NULL | NULL     | Using temporary |
+------+--------------+------------+------------+-------+---------------+--------------+---------+-------+------+----------+-----------------+

DERIVED：用來表示包含在FROM子句中的子查詢，mysql會遞歸執行并将結果放到一個臨時表中。伺服器内部稱為"派生表"，因為該臨時表是從子查詢中派生出來的。（派生表的性能不好）

--  MySQL 5.7開始優化器引入derived_merge，
-- 當子查詢中存在 UNION、GROUP BY、DISTINCT、LIMIT/OFFSET以及聚合操作時會禁止該優化
mysql> EXPLAIN SELECT * FROM ( SELECT * FROM zc_order LIMIT 1) temp;
+----+-------------+------------+------------+--------+---------------+------+---------+------+------+----------+-------+
| id | select_type | table      | partitions | type   | possible_keys | key  | key_len | ref  | rows | filtered | Extra |
+----+-------------+------------+------------+--------+---------------+------+---------+------+------+----------+-------+
|  1 | PRIMARY     | <derived2> | NULL       | system | NULL          | NULL | NULL    | NULL |    1 |      100 | NULL  |
|  2 | DERIVED     | zc_order   | NULL       | ALL    | NULL          | NULL | NULL    | NULL | 3781 |      100 | NULL  |
+----+-------------+------------+------------+--------+---------------+------+---------+------+------+----------+-------+

DEPENDENT DERIVED：表示派生表依賴于另一個表

SUBQUERY和UNION還可以被标記為DEPENDENT和UNCACHEABLE，DEPENDENT意味着SELECT在子查詢中并且依賴于外層查詢。UNCACHEABLE意味着無法緩存其結果的子查詢，必須為外部查詢的每一行重新計算其結果。

mysql> EXPLAIN SELECT * from zc_order where order_no in 
(SELECT order_no from zc_order o where o.order_no = 'UNO200418000000008' UNION SELECT order_no from zc_order o2 where o2.order_no = 'UNO200418000000009');
+------+--------------------+------------+------------+-------+---------------+--------------+---------+-------+------+----------+-----------------+
| id   | select_type        | table      | partitions | type  | possible_keys | key          | key_len | ref   | rows | filtered | Extra           |
+------+--------------------+------------+------------+-------+---------------+--------------+---------+-------+------+----------+-----------------+
|    1 | PRIMARY            | zc_order   | NULL       | ALL   | NULL          | NULL         | NULL    | NULL  | 3781 |      100 | Using where     |
|    2 | DEPENDENT SUBQUERY | o          | NULL       | const | IDX_ORDER_NO  | IDX_ORDER_NO | 130     | const |    1 |      100 | Using index     |
|    3 | DEPENDENT UNION    | o2         | NULL       | const | IDX_ORDER_NO  | IDX_ORDER_NO | 130     | const |    1 |      100 | Using index     |
| NULL | UNION RESULT       | <union2,3> | NULL       | ALL   | NULL          | NULL         | NULL    | NULL  | NULL | NULL     | Using temporary |
+------+--------------------+------------+------------+-------+---------------+--------------+---------+-------+------+----------+-----------------+

MATERIALIZED 物化子查詢，用于實作對子查詢的優化，對應FORMAT=JSON輸出的 materialized_from_subquery 詳見：MySQL官方文檔-subquery-materialization

table

表示該行所通路的表

<derivedN>

表示目前查詢依賴于id=N的查詢

<unionM,N>

表示id=M、N的查詢參與了UNION

<subqueryN>

id=N的物化子查詢結果

partitions

如果查詢是基于分區表的話，會顯示查詢将通路的分區

type

表示MySQL在表中找到所需行的方式，又稱“通路類型”，常見類型如下:

現有索引

ALTER TABLE zc_order ADD PRIMARY KEY (ID);

ALTER TABLE zc_order ADD UNIQUE INDEX IDX_ORDER_NO (ORDER_NO);

ALTER TABLE zc_order ADD INDEX IDX_USER_NO_ORDER_STATUS (USER_NO, ORDER_STATUS);

all: Full Table Scan， MySQL将周遊全表以找到比對的行

mysql> EXPLAIN SELECT * FROM zc_order;
+----+-------------+----------+------------+------+---------------+------+---------+------+------+----------+-------+
| id | select_type | table    | partitions | type | possible_keys | key  | key_len | ref  | rows | filtered | Extra |
+----+-------------+----------+------------+------+---------------+------+---------+------+------+----------+-------+
|  1 | SIMPLE      | zc_order | NULL       | ALL  | NULL          | NULL | NULL    | NULL | 3781 |      100 | NULL  |
+----+-------------+----------+------------+------+---------------+------+---------+------+------+----------+-------+

index: Full Index Scan，index與ALL差別為index類型隻周遊索引樹

mysql> EXPLAIN SELECT id, order_no FROM zc_order;
+----+-------------+----------+------------+-------+---------------+--------------+---------+------+------+----------+-------------+
| id | select_type | table    | partitions | type  | possible_keys | key          | key_len | ref  | rows | filtered | Extra       |
+----+-------------+----------+------------+-------+---------------+--------------+---------+------+------+----------+-------------+
|  1 | SIMPLE      | zc_order | NULL       | index | NULL          | IDX_ORDER_NO | 130     | NULL | 3781 |      100 | Using index |
+----+-------------+----------+------------+-------+---------------+--------------+---------+------+------+----------+-------------+

range: 索引範圍掃描，顯而易見的索引範圍掃描是帶有

between、!=、<>、in、not in、or

的查詢（or兩邊必須都為索引）

mysql> EXPLAIN SELECT * FROM zc_order WHERE order_no = 'UNO200418000000008' or order_no = 'UNO200418000000009';
+----+-------------+----------+------------+-------+---------------+--------------+---------+------+------+----------+-----------------------+
| id | select_type | table    | partitions | type  | possible_keys | key          | key_len | ref  | rows | filtered | Extra                 |
+----+-------------+----------+------------+-------+---------------+--------------+---------+------+------+----------+-----------------------+
|  1 | SIMPLE      | zc_order | NULL       | range | IDX_ORDER_NO  | IDX_ORDER_NO | 130     | NULL |    2 |      100 | Using index condition |
+----+-------------+----------+------------+-------+---------------+--------------+---------+------+------+----------+-----------------------+

ref: 使用非唯一索引掃描或者唯一索引的字首掃描（組合索引），傳回比對某個單獨值的記錄行

-- 非唯一索引，走ref
-- 唯一索引，如果是單列索引走的是const, 如果是組合索引的字首比對走ref
mysql> EXPLAIN SELECT * FROM zc_order WHERE USER_NO = 'US0000000001';
+----+-------------+----------+------------+------+--------------------------+--------------------------+---------+-------+------+----------+-------+
| id | select_type | table    | partitions | type | possible_keys            | key                      | key_len | ref   | rows | filtered | Extra |
+----+-------------+----------+------------+------+--------------------------+--------------------------+---------+-------+------+----------+-------+
|  1 | SIMPLE      | zc_order | NULL       | ref  | IDX_USER_NO_ORDER_STATUS | IDX_USER_NO_ORDER_STATUS | 258     | const |    4 |      100 | NULL  |
+----+-------------+----------+------------+------+--------------------------+--------------------------+---------+-------+------+----------+-------+

eq_ref: 類似ref，差別就在使用的索引是主鍵索引或者唯一索引，對于每個索引鍵值，表中隻有一條記錄比對

-- 單表沒測試出來，通常是const
-- 連表查詢
mysql> EXPLAIN SELECT * FROM zc_order o, zc_order_goods g WHERE o.ORDER_NO = g.ORDER_NO;
+----+-------------+-------+------------+--------+---------------+--------------+---------+----------------------+------+----------+-------+
| id | select_type | table | partitions | type   | possible_keys | key          | key_len | ref                  | rows | filtered | Extra |
+----+-------------+-------+------------+--------+---------------+--------------+---------+----------------------+------+----------+-------+
|  1 | SIMPLE      | g     | NULL       | ALL    | NULL          | NULL         | NULL    | NULL                 | 4696 |      100 | NULL  |
|  1 | SIMPLE      | o     | NULL       | eq_ref | IDX_ORDER_NO  | IDX_ORDER_NO | 130     | test-db.g.ORDER_NO   |    1 |      100 | NULL  |
+----+-------------+-------+------------+--------+---------------+--------------+---------+----------------------+------+----------+-------+

const: 當MySQL對查詢某部分進行優化，并轉換為一個常量時，使用這些類型通路。例如使用主鍵或唯一索引進行查詢

mysql> EXPLAIN SELECT * FROM zc_order WHERE order_no = 'UNO200418000000008';
+----+-------------+----------+------------+-------+---------------+--------------+---------+-------+------+----------+-------+
| id | select_type | table    | partitions | type  | possible_keys | key          | key_len | ref   | rows | filtered | Extra |
+----+-------------+----------+------------+-------+---------------+--------------+---------+-------+------+----------+-------+
|  1 | SIMPLE      | zc_order | NULL       | const | IDX_ORDER_NO  | IDX_ORDER_NO | 130     | const |    1 |      100 | NULL  |
+----+-------------+----------+------------+-------+---------------+--------------+---------+-------+------+----------+-------+

system: system是const類型的特例，當查詢的表隻有一行的情況下，使用system

NULL: MySQL在優化過程中分解語句，執行時甚至不用通路表或索引，例如從一個索引列裡選取最小值可以通過單獨索引查找完成

mysql> EXPLAIN SELECT MIN(order_no) FROM zc_order;
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+------------------------------+
| id | select_type | table | partitions | type | possible_keys | key  | key_len | ref  | rows | filtered | Extra                        |
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+------------------------------+
|  1 | SIMPLE      | NULL  | NULL       | NULL | NULL          | NULL | NULL    | NULL | NULL | NULL     | Select tables optimized away |
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+------------------------------+

從上到下，性能越來越好

possible_keys

指出MySQL能使用哪個索引在表中找到記錄，查詢涉及到的字段上若存在索引，則該索引将被列出，但不一定被查詢使用

有可能出現

possible_keys

不為null，但是

key

為null的情況，一般是查詢條件上有符合的索引，但是mysql優化器認為走全表掃描效率更高

mysql> EXPLAIN SELECT * FROM zc_order where user_no > 'US0000000013';
+----+-------------+----------+------------+------+--------------------------+------+---------+------+------+----------+-------------+
| id | select_type | table    | partitions | type | possible_keys            | key  | key_len | ref  | rows | filtered | Extra       |
+----+-------------+----------+------------+------+--------------------------+------+---------+------+------+----------+-------------+
|  1 | SIMPLE      | zc_order | NULL       | ALL  | IDX_USER_NO_ORDER_STATUS | NULL | NULL    | NULL | 3781 |    90.27 | Using where |
+----+-------------+----------+------------+------+--------------------------+------+---------+------+------+----------+-------------+

另一總情況剛好相反，possible_keys為null，但是key不為null，通常是使用了覆寫索引導緻的

-- 這裡比較有趣的一點是，覆寫的索引為id，任意索引中都存在主鍵 mysql根據某種規則選擇了其中一個而不是直接拿主鍵索引
-- 如果查詢的是user_no，那麼必定是使用IDX_USER_NO_ORDER_STATUS索引
mysql> EXPLAIN SELECT id FROM zc_order;
+----+-------------+----------+------------+-------+---------------+--------------+---------+------+------+----------+-------------+
| id | select_type | table    | partitions | type  | possible_keys | key          | key_len | ref  | rows | filtered | Extra       |
+----+-------------+----------+------------+-------+---------------+--------------+---------+------+------+----------+-------------+
|  1 | SIMPLE      | zc_order | NULL       | index | NULL          | IDX_ORDER_NO | 130     | NULL | 3781 |      100 | Using index |
+----+-------------+----------+------------+-------+---------------+--------------+---------+------+------+----------+-------------+

key

實際用到的索引

key_len

表示索引中使用的位元組數，可通過該列計算查詢中使用的索引的長度（key_len顯示的值為索引字段的最大可能長度，并非實際使用長度，即key_len是根據表定義計算而得，不是通過表内檢索出的）

所有的索引字段，如果沒有設定not null，則需要加一個位元組用于記錄是否為NULL
定長字段，int占四個位元組、date占三個位元組、char(n)占n個字元，tinyint占一個位元組
對于變長字段varchar(n)，則有n個字元+兩個位元組。
不同的字元集，一個字元占用的位元組數不同。latin1編碼的一個字元占用一個位元組，gbk編碼的一個字元占用兩個位元組，utf8編碼的一個字元占用三個位元組，utf8mb4編碼的一個字元占用四個位元組

-- `ORDER_NO` varchar(32) CHARACTER SET utf8 NOT NULL DEFAULT ''
-- 32 * 3 + 2 = 98
mysql> EXPLAIN SELECT * FROM zc_order where order_no = 'UNO200515000000060';
+----+-------------+----------+------------+-------+---------------+--------------+---------+-------+------+----------+-------+
| id | select_type | table    | partitions | type  | possible_keys | key          | key_len | ref   | rows | filtered | Extra |
+----+-------------+----------+------------+-------+---------------+--------------+---------+-------+------+----------+-------+
|  1 | SIMPLE      | zc_order | NULL       | const | IDX_ORDER_NO  | IDX_ORDER_NO | 98      | const |    1 |      100 | NULL  |
+----+-------------+----------+------------+-------+---------------+--------------+---------+-------+------+----------+-------+

-- `USER_NO` varchar(64) CHARACTER SET utf8 NOT NULL DEFAULT ''
-- `ORDER_STATUS` char(2) CHARACTER SET utf8 NOT NULL DEFAULT ''
-- 64 * 3 + 2 + 2 * 3 = 200
mysql> EXPLAIN SELECT * FROM zc_order where user_no = 'US0000000013' and order_status = '01';
+----+-------------+----------+------------+------+--------------------------+--------------------------+---------+-------------+------+----------+-------+
| id | select_type | table    | partitions | type | possible_keys            | key                      | key_len | ref         | rows | filtered | Extra |
+----+-------------+----------+------------+------+--------------------------+--------------------------+---------+-------------+------+----------+-------+
|  1 | SIMPLE      | zc_order | NULL       | ref  | IDX_USER_NO_ORDER_STATUS | IDX_USER_NO_ORDER_STATUS | 200     | const,const |    1 |      100 | NULL  |
+----+-------------+----------+------------+------+--------------------------+--------------------------+---------+-------------+------+----------+-------+

ref

表示索引的哪些列被使用了，可能是一個常量

常見的有：const、字段名、NULL

-- const
mysql> EXPLAIN SELECT * FROM zc_order WHERE id = 20;
+----+-------------+----------+------------+-------+---------------+---------+---------+-------+------+----------+-----------------------------------+
| id | select_type | table    | partitions | type  | possible_keys | key     | key_len | ref   | rows | filtered | Extra                             |
+----+-------------+----------+------------+-------+---------------+---------+---------+-------+------+----------+-----------------------------------+
|  1 | SIMPLE      | zc_order | NULL       | const | PRIMARY       | PRIMARY | 8       | const |    1 |      100 | Directly search via Primary Index |
+----+-------------+----------+------------+-------+---------------+---------+---------+-------+------+----------+-----------------------------------+

-- 字段名
mysql> EXPLAIN SELECT * FROM zc_order o, zc_order_goods g WHERE o.ORDER_NO = g.ORDER_NO;
+----+-------------+-------+------------+--------+---------------+--------------+---------+----------------------+------+----------+-------+
| id | select_type | table | partitions | type   | possible_keys | key          | key_len | ref                  | rows | filtered | Extra |
+----+-------------+-------+------------+--------+---------------+--------------+---------+----------------------+------+----------+-------+
|  1 | SIMPLE      | g     | NULL       | ALL    | NULL          | NULL         | NULL    | NULL                 | 4696 |      100 | NULL  |
|  1 | SIMPLE      | o     | NULL       | eq_ref | IDX_ORDER_NO  | IDX_ORDER_NO | 130     | test-db.g.ORDER_NO   |    1 |      100 | NULL  |
+----+-------------+-------+------------+--------+---------------+--------------+---------+----------------------+------+----------+-------+

rows

表示MySQL根據表統計資訊及索引選用情況，估算的找到所需的記錄所需要讀取的行數

filtered

這個字段表示存儲引擎傳回的資料在server層過濾後，剩下多少滿足查詢的記錄數量的百分比（估計）

這個字段幾乎沒什麼用，通常都是顯示100%，隻對索引類型為

index

和

all

時才生效（where條件中出現的列上建有索引，但是執行計劃不走索引的range或ref掃描，而走全表掃描或覆寫索引掃描）

mysql> EXPLAIN SELECT * FROM zc_order where user_no > 'US0000000013';
+----+-------------+----------+------------+------+--------------------------+------+---------+------+------+----------+-------------+
| id | select_type | table    | partitions | type | possible_keys            | key  | key_len | ref  | rows | filtered | Extra       |
+----+-------------+----------+------------+------+--------------------------+------+---------+------+------+----------+-------------+
|  1 | SIMPLE      | zc_order | NULL       | ALL  | IDX_USER_NO_ORDER_STATUS | NULL | NULL    | NULL | 3755 |    49.99 | Using where |
+----+-------------+----------+------------+------+--------------------------+------+---------+------+------+----------+-------------+
1 row in set

Extra

包含不适合在其他列中顯示但十分重要的額外資訊

Using where

表示mysql伺服器将在存儲引擎檢索後再進行過濾（service層過濾）。許多where條件裡涉及索引中的列，當它讀取索引時，能夠直接在索引中進行過濾，是以不是所有帶where的查詢都會顯示"Using where"
Using index

該值表示相應的select操作中使用了覆寫索引（Covering Index）
Using index confition

表示使用到了索引下推

【MYSQL】EXPLAIN指令EXPLAIN指令idselect_typetablepartitionstypepossible_keyskeykey_lenrefrowsfilteredExtra

Using temporary

表示MySQL需要使用臨時表來存儲結果集，常見于排序和分組查詢
Using filesort

MySQL必須做一個額外的步驟來找出如何按排序的順序檢索行
Using join buffer

該值強調了在擷取連接配接條件時沒有使用索引，并且需要連接配接緩沖區來存儲中間結果。如果出現了這個值，那應該注意，根據查詢的具體情況可能需要添加索引來改進能
Impossible where

這個值強調了where語句會導緻沒有符合條件的行

Select tables optimized away

這個值意味着僅通過使用索引，優化器可能僅從聚合函數結果中傳回一行。

+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+------------------------------+
| id | select_type | table | partitions | type | possible_keys | key  | key_len | ref  | rows | filtered | Extra                        |
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+------------------------------+
|  1 | SIMPLE      | NULL  | NULL       | NULL | NULL          | NULL | NULL    | NULL | NULL | NULL     | Select tables optimized away |
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+------------------------------+

Index merges

當MySQL 決定要在一個給定的表上使用超過一個索引的時候，就會出現以下格式中的一個，詳細說明使用的索引以及合并的類型

Using sort_union(…)

Using union(…)

Using intersect(…)

【MYSQL】EXPLAIN指令EXPLAIN指令idselect_typetablepartitionstypepossible_keyskeykey_lenrefrowsfilteredExtra

EXPLAIN指令

id

select_type

table

partitions

type

possible_keys

key

key_len

ref

rows

filtered

Extra

繼續閱讀

2022秋招面試總結（cpp+java+測開）百度測開一面位元組後端一面蝦皮後端一面蝦皮後端二面

資料庫之DDL操作資料庫DDL操作資料庫DDL操作資料表

資料庫之DQL操作資料庫

mysql優化（sql優化）

資料遷移方法資料遷移原則資料遷移之雙寫方案資料遷移之級聯同步方案

redis叢集資料一緻性_RedisRaft為Redis叢集帶來強大的資料一緻性

寶塔面闆mysql恢複2018.1.8更新

Centos7 MySQL 5.7 安裝MySQL 5.7 安裝

查找入職員工時間排名倒數第三的員工所有資訊

Hibernate使用Hibernate的“3個準備，7個步驟”Hibernate API簡介操作實體對象對象識别

雲計算面試題——mysql/存儲引擎/備份

SQL語言基礎：常用的資料查詢語句

Ubuntu16.04安裝Apache+MySQL+PHP1. 安裝Apache2. 安裝MySQL3. 安裝PHP4. 安裝phpMyAdmin

MySQL的4種隔離級别？出現問題

neo4j之cypher使用文檔

mysql使用source指令導入.sql檔案