Redis系列（七）：緩存只是讀寫回種這麼簡單嗎？如果是，那麼請你一定看看這篇文章！

{"type":"doc","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"  "}]},{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"一枚用心堅持寫原創的“無趣”程序猿，在自身受益的同時也讓朋友們在技術上有所提升。"}]}]},{"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":"前面利用 6 篇文章講述了 Redis 相關的基礎知識，相信小夥伴們對 Redis 已經有了一個比較深入的認識和理解了；本文來講講實際生產環境中 Redis 作爲常用緩存組件是怎麼和 DB（關係型數據庫，比如 MySQL）配合使用的。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"看到這裏可能有些朋友會內心肯定會淡淡的說上一句："},{"type":"text","marks":[{"type":"strong"}],"text":"寫操作先更新 DB，然後在更新緩存，讀操作先讀緩存，如果沒有讀 DB 回種緩存，然後返回結果不就完事了麼"},{"type":"text","text":"，這有什麼好講的？"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"只要你有這樣的疑問，那麼請你一定認真看完本文，因爲緩存讀寫策略遠不止你想的那麼簡單，下面我們就來分析一下幾種常用的緩存讀寫方案，看看哪種更適合你。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"讀寫緩存策略一"}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/18/1869454a63c26a13f4cf7ddec0c126c4.png","alt":null,"title":"","style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"boxShadow"}],"href":"","fromPaste":false,"pastePass":false}},{"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":"如上圖所示，對於寫操作，先寫 DB，如果 DB 成功，則同步寫入緩存；對於讀操作，首先從緩存中讀取，若緩存未命中，則從 DB 中獲取，如果取到了結果，將結果回種緩存並返回，若 DB 中也沒有結果，則在緩存中設置一個短暫帶有過期時間的空值，防止相同 key 頻繁請求對 DB 造成大量的無效請求。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"策略一優缺點對比"}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"優點"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"緩存讀寫策略簡單（在很多讀多寫少的場景中，此種策略使用的頻率還是很高的）。"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"不需要依賴其他第三方緩存組件協同完成。"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"缺點"}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"zerowidth"}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/90/904029ca6ae8a6225c8170e079333fb6.png","alt":null,"title":"","style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"boxShadow"}],"href":"","fromPaste":false,"pastePass":false}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"舉個例子，如上圖所示，假設兩條線程 1 和線程 2 同時對一件特價商品進行價格調整，假設商品初始狀態爲 20 元，線程 1 將其價格修改爲 19 元並寫入緩存，然後線程 2 將商品的價格修改爲 18 元，此時線程 A 進行讀取操作，因爲線程 2 還沒來得及將緩存數據修改爲 18，所以此時線程 1 拿到的價格爲 19 元，但是此時庫裏的價格卻是 18 元。從這個例子可以看出，對於進行頻繁修改的數據，使用此種緩存讀寫策略顯然是不合理的。針對這種緩存讀寫方案的缺陷，我們來看看下一種緩存讀寫策略。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Cache Aside 讀寫策略"}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/35/3594adb25dfd791d92623df511f4f2d2.png","alt":null,"title":"","style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"boxShadow"}],"href":"","fromPaste":false,"pastePass":false}},{"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":"Cache Aside 讀寫策略也叫緩存旁路讀寫策略，從上圖可以看出，針對讀寫的策略分別是："}]},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"讀流程："}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"首先從緩存中讀取，如果有結果則直接返回結束"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"如果緩存中沒有，則讀 DB 並將結果回種緩存，然後返回結束"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"寫流程："}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"首先將將結果寫入 DB"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"然後刪除緩存中對應的值"}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"zerowidth"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"實際生產環境中，此種策略的使用也相對比較廣泛，可以作爲一種參考。這裏需要注意一點的是，針對寫流程，不能先刪除緩存，在更新 DB，因爲緩存刪除後，此時 DB 還沒有更新完時，來了一個 get 請求，那麼緩存就有可能會被種入一個已失效的結果。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Cache Aside 讀寫策略優缺點對比"}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"優點："}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"從流程圖看，此種策略實現比較簡單。"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"對於緩存和 DB 的一致性有了一定的保證，其可以解決第一種緩存方案遇到的問題。"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"缺點："}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"很顯然，每次更新數據，都會先更新 DB 緊接着就刪除緩存，如果讀寫操作都比較頻繁的情況下，勢必使得緩存的命中率有所折扣，也就意味着緩存的 miss 率升高，從而導致在一定程度上削弱了緩存的作用。"}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"針對此種方案的缺點，其實也有一些比較折中的方案可以考慮。比如在更新 DB 完成後，同樣更新緩存，但是在更新緩存的時候增加分佈式鎖避免；在比如如果業務場景並不是要求強一致性的話，可以將數據寫入緩存並增加一個過期時間，這樣即使數據不一致也只是一段時間。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"對於增加過期時間的這種方式，存在極熱 key 的場景並不是適用的，因爲一旦 key 過期後，在一瞬間大量請求越過緩存，直衝 DB，也就造成了緩存穿透的問題，所以 Cache Aside 方案看起來很不錯，但是也不是萬能的。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Write/Read Through 策略"}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/50/500c4be18059b6fdeb9d775422ec0012.png","alt":null,"title":"","style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"boxShadow"}],"href":"","fromPaste":false,"pastePass":false}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Write/Read Through 的緩存策略不同於前兩種，該策略需要引入第三方緩存同步插件。其讀寫流程如下："}]},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"讀流程："}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"首先讀緩存，如果緩存命中，則直接返回結果"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"如果緩存未命中，則依賴第三方組件從 DB 中加載數據到緩存中，然後將獲取的結果返回。"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"寫流程："}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"要更新的數據是否在緩存中存在，若存在則直接將數據寫入緩存，之後緩存數據由第三方緩存組件將其更新到 DB"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"若緩存中不存在，則直接將結果寫入 DB，這種稱之爲寫穿透"}]}]}]},{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"zerowidth"}]},{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Write/Read Through 策略優缺點對"}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"優點："}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"寫緩存 miss 的情況除外，剩餘所有操作都只與緩存進行交互，很大程度上避免了緩存與 DB 一直性的情況"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"缺點："}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"從上面流程中不難看出，寫緩存 miss 時直接與 DB 交互，會造成請求耗時增加"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"此種緩存策略引入了第三方緩存組件進行輔助，從而增加了系統的複雜度和系統的維護難度"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"由於需要引入第三方組件，而目前很多緩存如 Redis 原生並不兼容第三方組件，所以很難引入"}]}]}]},{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"zerowidth"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"總和上面這些情況對比，目前採用此種緩存讀寫策略的場景很少，作者到目前還沒有見過使用這種緩存策略的場景。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Write Back 策略"}]}]},{"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":"Write Back 策略是一種操作系統在使用的緩存讀寫策略，由於其實現比較複雜，所以讀者朋友可以做一些瞭解即可，目前沒有在生產環境中實際使用過。"}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/b5/b5da8cf1210f5c73d021b62040ee20ec.png","alt":null,"title":"","style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"boxShadow"}],"href":"","fromPaste":false,"pastePass":false}},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"寫操作流程如下："}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"寫請求來之後，首先判斷要寫的 key 在緩存中是否被標記爲“髒數據”，如果不是“髒數據”則直接寫緩存，然後將其標記爲髒數據"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"如果緩存中標記的是“髒數據”，則直接將其寫入 DB，然後回種緩存，然後將其標記爲“髒數據”"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"讀操作流程如下："}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"首先從緩存中獲取數據，若有則直接返回"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"若緩存中沒有，則尋找可用的緩存快，若緩存快被標記爲“髒數據”，則將“髒數據”寫入 DB，然後將緩存的數據標記爲“非髒數據”，然後返回"}]}]},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"若緩存快中的數據不是“髒數據”，則從 DB 中加載數據到緩存中，然後將其返回。"}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"生產環境中其他的緩存讀寫策略"}]}]},{"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":"除了上面我們介紹的 4 中緩存讀寫策略，實際生產環境中還有一些比較常見的策略，比如針對熱點 key 使用的 LRU 緩存策略。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"實際生產中，某些場景數據量是非常巨大的，比如微博用戶 uid，方便查看某個用戶的最近狀態，如果全部將其全部都寫入到緩存，顯然是不合理的；一是緩存存儲不了那麼多用戶信息，二是對應絕大部分用戶是完全沒有必要寫入緩存的，因爲對於很大一部分用戶信息，只有很少的訪問量。所以對於這種場景，只需要緩存最近經常被訪問的那部分用戶信息即可，針對這種場景，也就誕生了 LRU 緩存。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"其實 LRU 緩存在很多框架中也被廣泛使用，需要的朋友可以自己研究下很簡單的（這裏其實也有有一道算法題，感興趣的朋友可以自行在 LeetCode 上搜索）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"總結"}]}]},{"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":"本文介紹了 5 中生產環境中經常用到的緩存讀寫策略，讀者朋友們可以根據自己的實際業務場景，對其進行適當的改造就可以應用到自己的環境中了。好了，就講這麼多，更多的需要朋友們自行多多體會和應用。下篇文章見，拜拜了您。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/a3/a3f4e6f6c743badafb9e7c13e014c698.jpeg","alt":null,"title":"","style":[{"key":"width","value":"25%"},{"key":"bordertype","value":"boxShadow"}],"href":"","fromPaste":false,"pastePass":false}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"  "}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"往期推薦"}]},{"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://mp.weixin.qq.com/s?__biz=MzIzNTIzNzYyNw==&mid=2247484290&idx=1&sn=9e4d47ba8042543c9718ebcb766c8df3&chksm=e8eb7a16df9cf30057e2ccbc224f336a32b92b3342c461c8d54572913bb9cbac095266f5e4d4&scene=21#wechat_redirect","title":null},"content":[{"type":"text","text":"Redis系列（三）：緩存過期該如何剔除？RDB和AOF又是什麼？"}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"http://mp.weixin.qq.com/s?__biz=MzIzNTIzNzYyNw==&mid=2247483919&idx=1&sn=ee4c393dabadcecd51b34380425d227a&chksm=e8eb7b9bdf9cf28da20a8083d772d6744850256e7b338aa79225889c69f8395ab6019808d946&scene=21#wechat_redirect","title":null},"content":[{"type":"text","text":"你必須要知道集羣內部工作原理的一些事！"}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"http://mp.weixin.qq.com/s?__biz=MzIzNTIzNzYyNw==&mid=2247483881&idx=1&sn=6624157a86dea60abd8a6842c863ba4e&chksm=e8eb787ddf9cf16b85561dcd25144a73304709a14f615291a5374f9ca2bf05a2284b181be1af&scene=21#wechat_redirect","title":null},"content":[{"type":"text","text":"消息是如何在服務端存儲與讀取的，你真的知道嗎？"}]},{"type":"text","text":" "}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"http://mp.weixin.qq.com/s?__biz=MzIzNTIzNzYyNw==&mid=2247483827&idx=1&sn=44cfb50953b0f745b2719977453c3a42&chksm=e8eb7827df9cf131ea8aa1496e9ec2d1ed22ed245225dad0613511006b9428fffb16c92cdd6a&scene=21#wechat_redirect","title":null},"content":[{"type":"text","text":"一文讀懂消費者背後的那點\"貓膩\""}]}]}]}