到底什麼時候要分庫分表？

{"type":"doc","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"http://shardingsphere.apache.org/index_zh.html","title":"","type":null},"content":[{"type":"text","text":"http://shardingsphere.apache.org/index_zh.html","attrs":{}}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/a2/a2b697684caa03b624efb2a1507e4e2a.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"本質上：分佈式的數據中間件--解決方案；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"https://shardingsphere.apache.org/document/current/cn/overview/快速指南：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"主要的事兩個東西：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":" shardingsphere-JDBC輕量級 Java 框架，在 Java 的 JDBC 層提供的額外服務。 ","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"它使用客戶端直連數據庫，以 jar 包形式提供服務，無需額外部署和依賴，可理解爲增強版的 JDBC 驅動，完全兼容 JDBC 和各種 ORM 框架。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"shardingSphere -proxy:","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"透明化的數據庫代理端","attrs":{}},{"type":"text","text":"，提供封裝了數據庫二進制協議的服務端版本，用於完成對異構語言的支持","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/35/3527f0d855393c7badb43684a9a70993.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"爲什麼會出現這個數據庫中間件的軟件：shardingSphere 的數據庫中間件","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"codeblock","attrs":{"lang":"text"},"content":[{"type":"text","text":" 在電商環境中，我們一般是對於數據量是在逐步提升的一個狀態，在數據庫中的量級達到百萬級別或者是千萬級別之後，數據庫中的基本操作CRUD是比較耗費時間的，可能會導致用戶體驗降低， 雖然我們可以通過 ：\n ①sql優化\n ②索引優化\n ③分庫分表\n","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"當然我們也可以從硬件上提高服務器的存儲水平和處理業務的效率，增加硬盤的存儲空間和增加內存等，但是這些都是暫時可以提供一些便利；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"比如十年前的jd:數據庫的db設計:","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"表示的是簡單dB中，可以將數據單表的mysql中的數據表，單表的存儲量","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/3e/3e9aebb8509ebccd409e75828580c350.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"隨時物流，網絡時代，和個人消費水平的提高，電商平臺買東西已經是家常便飯，但是這樣的話，要求平臺數據性能在快的同時，也對數據庫的效率提出了更高的要求，比如數據要分庫分表，所謂分庫：","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/c4/c42249ad9c7f9b3602e1665edcab8c65.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"注意事項：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"我們這裏顯示的事主要的對比結果；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"mysql的單表保持在多少比較合適，前百度DBA中流傳出來的，當mysql表中的單表的數據量達到2000萬行時，sql的性能直線下降不久前，阿里提出的開發手冊中提到單表數據超過","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"500萬行","attrs":{}},{"type":"text","text":"或者單表的容量超過","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"2G","attrs":{}},{"type":"text","text":"，","attrs":{}},{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"考慮分庫分表","attrs":{}},{"type":"text","text":"；，顯然我們沒法測試具體的mysql的數據表瓶頸，理論上是T級別的，但是可能會隨着操作系統的文件的管控而變化；阿里鐵律造就額這個數據；","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/81/81b1df472e31b8009f6b98d4b5d47230.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"3.sql的編寫需要注意優化","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用limit對查詢結果的記錄進行限定","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":" 避免select *，將需要查找的字段列出來","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用連接（join）來代替子查詢","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"拆分大的delete或insert語句","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"可通過開啓慢查詢日誌來找出較慢的SQL","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"不做列運算：SELECT id WHERE age + 1 = 10，任何對列的操作都將導致表掃描，它包括數據庫教程函數、計算表達式等等，查詢時要儘可能將操作移至等號右邊","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"sql語句儘可能簡單：一條sql只能在一個cpu運算；大語句拆小語句，減少鎖時間；一條大sql可以堵死整個庫","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"OR改寫成IN：OR的效率是n級別，IN的效率是log(n)級別，in的個數建議控制在200以內","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"不用函數和觸發器，在應用程序實現","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"避免%xxx式查詢","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"少用JOIN","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用同類型進行比較，比如用'123'和'123'比，123和123比","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"儘量避免在WHERE子句中使用!=或<>操作符，否則將引擎放棄使用索引而進行全表掃描","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"對於連續數值，使用BETWEEN不用IN：SELECT id FROM t WHERE num BETWEEN 1 AND 5","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"列表數據不要拿全表，要使用LIMIT來分頁，每頁數量也不要太大","attrs":{}}]}]}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"這時候我們需要的的sharding sphere的數據庫中間件就大有用處了","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"4.垂直分表：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"將一個表中經過邏輯拆分成兩個表一個存放具體的基本信息，用於列表展示，一個存放特殊描述信息，用於點擊之後具體的課程信息；","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/c6/c6e84cd1cd20578b08d42de2761c0ec9.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"這就有很多好處，如果說我們需要改變課程描述的信息，但是在未垂直分表之前，我們將數據是其他課程基本信息不能動的，但是現在是分表之後基本信息不影響；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"單數據庫多表的操作會使得性能耗費巨大，io增多；","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"5.垂直分庫：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"把具體的業務劃分出具體的數據庫，形成專表專庫的形式；","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/6e/6e280c2f528e0dbb8fa47e643c64ad9c.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"水平分庫：單表數據量大，單表中也不能進行數據字段的格式垂直切分；","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/85/8589d47f16f8e3fad2a617dfc52a59a9.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"垂直改變的事數據的結構","attrs":{}},{"type":"text","text":"， 表的結構分開，業務系統的專用數據庫","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"水平對於的是表中的數據來操作","attrs":{}},{"type":"text","text":"， 分流數據，分爲多個庫，庫中在細分","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/90/908ea6ec0e7065ce9b114c11fe851227.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":" 總結：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":" 對於業務系統的垂直和水平的分層，可以更好的整理庫中的字段關係，爲高併發下的生成數據打下基礎","attrs":{}}]}]}