架構師視角 | 分佈式緩存如何選擇 ?

{"type":"doc","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"如今，緩存系統的應用非常廣泛，能夠用來提高併發數、數據吞吐量，提高快速響應能力。那麼當數據量達到一定程度，單機環境可能就顯得有些力不從心了，就需要一個分佈式緩存系統。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":3},"content":[{"type":"text","text":"1. 緩存系統的選擇","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/d2/d2a6ab302e0ab5c7302ebfc95f76520f.png","alt":"","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":"​","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"1.1 緩存分類","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"如上圖所示，首先緩存大致可以分爲四大類。","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"CDN 緩存：CDN 即內容分發網絡，CDN 邊緣節點將數據緩存起來。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"反向代理緩存：如 Nginx 的緩存。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"本地緩存：代表的有 EhCache 和 Guava Cache。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"分佈式緩存：各緩存系統。","attrs":{}}]}],"attrs":{}}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"1.2 分佈式緩存","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"本文主要探討各分佈式緩存系統，如圖 1-1 所示，列出了五種：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"其中 EvCache 和 Aerospike 使用場景不是那麼通用和廣泛。","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"EvCache：是 Netflix 的基於 Memcached & Spymemcached 的緩存方案。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Aerospike：是可基於 SSD 的 KV NoSQL 數據庫。","attrs":{}}]}],"attrs":{}}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"除此之外，還有三種常見緩存系統。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/a8/a8beb9ad9ddbe9458aaeed256b0f7429.png","alt":"","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":"​","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Tair：阿里開源，跨機房、性能隨結點添加線性上升、適用大數據量。Tair 還有三種引擎。","attrs":{}}]},{"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","text":"LDB: 基於 google levelDB，支持 KV和類 HashMap 結構，性能稍低，持久化可靠性最高。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"MDB: 基於 Memcache，支持 KV 和類 HashMap，性能最優，不支持持久化存儲。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"RDB: 基於 Redis。","attrs":{}}]}],"attrs":{}}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Memcache: 不支持數據同步、分佈式支持較差。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Redis: 社區活躍、使用最多。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"綜上所述，在一般情況下，考慮到適用性和穩定性，Redis 是搭建緩存系統的最優選擇。以下將基於 Redis 介紹。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":3},"content":[{"type":"text","text":"2. Redis 集羣緩存方案","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"如頂部圖 1-1 所示，列出了 Redis 的集羣高可用的方案，基本可以分爲三種。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"2.1 主從機制","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"常見的集羣架構，搭建簡單，主要實現讀寫分離和備份，可以由 Master 負責讀寫，Slave 負責備份。但存在故障恢復複雜、水平拓展難、寫能力受限等問題。結構圖如下：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/b1/b136679b1303df46347c5a57ce0960e0.png","alt":"","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":"​","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"2.2 哨兵機制","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Redis Sentinel 是社區版本推出的原生高可用解決方案。由一或多個哨兵實例監視任意個主從服務器，且在 Master 宕機時，自動將宕機服務器屬下的 Slave 服務器升級爲 主服務器，從而保證系統的可用性。較主從實現的監控、選主。但問題主要是要保證 Master 的 HA 切換。結構圖如下：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/fa/fa33d8cafc52d2c4ecee64ad6c7058a6.png","alt":"","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":"​","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"2.3 \"分佈式\"","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"到這裏以上兩種機制其實只能算作“集羣”，並非嚴格意義上的“分佈式”。接着來看看分佈式方案。","attrs":{}}]},{"type":"blockquote","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"集羣強調高可用，分佈式在集羣的基礎上又強調協作。","attrs":{}}]}],"attrs":{}},{"type":"heading","attrs":{"align":null,"level":3},"content":[{"type":"text","text":"3. Redis分佈式緩存方案","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"任何分佈式存儲系統，首先面臨的就是 sharding（分片）問題，如頂部圖 1-1 所示該問題有爲三種解決方法。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"3.1 客戶端分片","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"顧名思義，將數據分片的路由功能交給客戶端，但這是一種靜態分片，維護性差。基本是不予考慮的。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"3.2 代理分片","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"通過代理分發到具體的 redis 實例。有兩個常用解決方案。","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Twemproxy：Twitter 開源，輕量級，不再維護，無法平滑地擴容/縮容，運維也不是很友好，性能一般。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Codis：豌豆莢開源，支持水平拓展，運維平臺完善，性能較 Twemproxy 快。Codis 在國內使用的較多，同時代理分片的思路也有很多公司在此基礎開發了自己的二次方案。不過 Codis 也不再維護。","attrs":{}}]}],"attrs":{}}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"其實，這兩種代理分片的方案，都是在 Redis 官方未推出良好的分佈式方案時的產生的，在官方更新提供更優策略後都不再維護。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"3.3 服務器端分片","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"這就要談到 Redis 官方方案 Redis-cluster 。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在 Redis 3.0 之前是沒有較好的分佈式方案的，這也是第三方方案出現的原因。3.0 開始，官方推出了去中心化的分佈式方案。集羣中包含 16384 個散列槽，每個節點負責其中一部分。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"先看下拓撲圖：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/49/493ce1fb5e40cb7b200f808db253c42d.png","alt":"","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":"​","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"每個節點打開兩個 TCP 連接，一個負責給客戶端提供服務，一個負責節點間通信。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"此刻要說說 CAP 了 ：Consistency（一致性）、Availability（可用性）、Partition tolerance（分區容錯性） 。對分佈式系統而言，CAP 必須犧牲一者。Redis Cluster 的設計目標主要是高性能、高可用和高擴展，只好拋棄一部分數據一致性。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"數據一致性：由於Redis Cluster 使用異步複製， 在某些情況下如 Master 宕機但未同步至 Slave，可能會導致丟失寫入。在絕對需要支持同步寫入時，可通過 WAIT 命令實現，可使得丟失寫入的可能性大大降低。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"可用性：當集羣中一部分節點故障後，集羣整體能響應客戶端讀寫請求。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/66/66b5ff020152927fae87c76dfbff218c.png","alt":"","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":"​","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"節點間定時互 ping ，當超過一半 Master 判定某節點失敗，則標記爲 FAIL，且會向集羣廣播節點下線的消息。如下線節點是帶有槽的主節點，則要從它的從節點選出一個替換。","attrs":{}}]}],"attrs":{}}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"高性能和拓展：操作某個 key 時，不會先找到節點再處理，而是直接直接重定向到該節點，同時相較代理分片也少了 proxy 的連接損耗。","attrs":{}}]},{"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","text":"但是在進行 multiple key 操作時需要 keys 位於同一個 slot 上，需要使用 hash tags，使用 {} 強制將某些 key 映射到每個 slot，以便進行 multiple 。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在拓展方面，Redis Cluster 最大支持線性拓展 1000 個節點，將新節點加入集羣后可以通過命令指定和平均的從已有節點分配 slot。","attrs":{}}]}],"attrs":{}}],"attrs":{}},{"type":"heading","attrs":{"align":null,"level":3},"content":[{"type":"text","text":"4. 緩存常見問題","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"以上介紹了簡單介紹了常見緩存系統，並具體列出了基於 Redis 的集羣方案。下面談一談緩存系統常見的問題。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"如下圖所示，列出七個常見問題。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/cc/ccdc0c31c295c9243fc144ed5b02d6a4.png","alt":"","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":"​","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"4.1. 緩存穿透","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"指訪問不存在的數據，從而繞過緩存，直接請求到了數據源，當請求過多，就會對 DB 造成壓力。","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"空 key：指對於不存在的數據也將 key 存空值入緩存系統，這樣下次訪問也會得到返回。但只適用於空數據 key 有限、key 重複請求概率高，如果量大且不重複，就會造成很多無用 key 的創建。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"布隆過濾器：布隆過濾器是一個很長的二進制向量和一系列隨機映射函數。可用於檢索一個元素是否在一個集合中加一層對空值的過濾器，空間和時間效率都很高。但由於 hash 產生的碰撞可能存在誤判，以及因不存儲 key 導致的無法刪除。適用於空數據 key 各不同、重複請求概率低。","attrs":{}}]}],"attrs":{}}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"4.2. 緩存擊穿","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"緩存擊穿實際是緩存雪崩的一個特例。指當某些熱點 key 過期時，就會有大量的請求擊穿到 DB。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"互斥鎖：在緩存失效的時候，不立即 load db，可以先用如 SETNX 等命令去 set 一個 mutex key，當操作返回成功時，說明拿到鎖，此刻該線程進行 load db 的操作並更新緩存；否則未拿到鎖就（可休眠一段）重試 get 緩存的方法。但要注意死鎖風險。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"不過期","attrs":{}}]},{"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","text":"這裏的不過期有兩個概念，一個指未設過期時間，那是真的不過期，那沒事了。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"另一個是指通過業務邏輯，將 key 的過期時間進行存儲，請求是判斷是否小於值，是則後臺異步更新。","attrs":{}}]}],"attrs":{}}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"4.3. 緩存雪崩","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"同一時刻大量緩存失效（故障）, 請求到了 DB。","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"隨機時間：在設置過期時間時，可以在基礎時間上 + 一個隨機的時間，等於實現了分批過期。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"後臺更新：將更新失效的工作交給後臺定時線程。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"限流 + 本地緩存：如 ehcache 本地緩存 + Hystrix 限流。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"雙緩存：類似於設置主從緩存，從 key 不過期。","attrs":{}}]}],"attrs":{}}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"4.4. 緩存更新與一致性","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"如果保證數據一致性。列出四種更新策略：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Cache Aside ：最常用的。失效時回源取數據，更新；命中時，返回緩存數據；更新時先數據源更新，再更新緩存。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Write Back ：更新數據時，只更新緩存，不更新數據源。緩存異步批量更新數據庫。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Read/Write Through","attrs":{}}]},{"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","text":"Write Through ：當有數據更新時，如未命中緩存，直接更新數據庫，並返回。如命中緩存，則更新緩存，再由 Cache 自己更新數據庫。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Read Through ：更新數據源由緩存系統操作，讀取數據時如緩存失效，則取回源數據更新緩存。","attrs":{}}]}],"attrs":{}}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"4.5. 熱點數據","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"對於熱點數據的處理方法。","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"拆分複雜結構：如二級數據結構，進行拆分，這樣熱點 key 就被拆爲若干個的 key 分佈到不同節點。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"遷移熱點：對於 Redis Cluster 而言可以將熱點 key 所在的 slot 單獨遷移到一個節點，降低其他節點壓力。","attrs":{}}]}],"attrs":{}},{"type":"listitem","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"多副本：複製多份緩存副本，將請求分散到多個節點上，減輕單臺緩存服務器壓力，適合多讀少寫。","attrs":{}}]}],"attrs":{}}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"4.6. 緩存預熱","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"指可以將某些的緩存數據提前加載到緩存系統，提前避免在如熱點數據大量請求到庫。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"4.7. 緩存降級","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"指當訪問量劇增、服務出現問題或非核心服務影響到核心流程的性能時，仍需保證主服務可用。可根據一些關鍵數據自動降級，也可配置開關人工降級。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":3},"content":[{"type":"text","text":"5. Redis Cluster 使用","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"對於 Redis Cluster 環境的搭建和基礎使用非常簡單。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"無論基於何種方式，只要搭建好 n 臺 redis 服務並保證各服務間可以互相通訊後，任意進入一個 redis 服務鍵入：","attrs":{}}]},{"type":"codeblock","attrs":{"lang":null},"content":[{"type":"text","text":"redis-cli --cluster create IP1:port1 IP2:port2 IP3:port3 IP4:port4 IP5:port5 IP6:port6 ... --cluster-replicas 1","attrs":{}}]},{"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":"即可。之後可以使用 cluster node 和 cluster info 命令查看集羣、節點信息。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"而對於廣大 JAVA 開發，Spring Data Redis 從 1.7 起即支持 Redis Cluster，只需配置 Master 節點地址（和密碼）。","attrs":{}}]},{"type":"codeblock","attrs":{"lang":null},"content":[{"type":"text","text":"spring.redis.cluster.nodes=ip1:port1,ip2:port2,ip3:port3","attrs":{}}]},{"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":"加入依賴","attrs":{}}]},{"type":"codeblock","attrs":{"lang":null},"content":[{"type":"text","text":"compile(\"org.springframework.boot:spring-boot-starter-data-redis\")","attrs":{}}]},{"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":"即可通過 RedisTemplate 使用。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":3},"content":[{"type":"text","text":"6. 總結","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"本文從緩存系統的選擇出發，基於 Redis 介紹了幾種集羣方案並重點說明了 Redis Cluster 方案。之後列出緩存系統常見問題及常見解決方案，最後對使用做了簡單說明。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"當然，如何去落地，如何解決這些問題還需要根據實際場景具體分析和處理。","attrs":{}}]},{"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":"link","attrs":{"href":"https://www.bilibili.com/video/BV17z4y1k7zF/","title":null},"content":[{"type":"text","text":"MySQL從入門到進階教程，主講老師：馬士兵、連鵬舉","attrs":{}}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https://blog.csdn.net/Java0258/article/details/109403289?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522160665848319195265186288%2522%252C%2522scm%2522%253A%252220140713.130102334.pc%255Fall.%2522%257D&request_id=160665848319195265186288&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2~all~first_rank_v2~rank_v28-1-109403289.pc_first_rank_v2_rank_v28&utm_term=%E5%AD%97%E8%8A%82%E8%B7%B3%E5%8A%A8%E6%80%BB%E7%BB%93%E7%9A%84%E8%AE%BE%E8%AE%A1%E6%A8%A1%E5%BC%8F%20PDF%20%E7%81%AB%E4%BA%86%EF%BC%8C%E5%AE%8C%E6%95%B4%E7%89%88%E5%BC%80%E6%94%BE%E5%88%86%E4%BA%AB&spm=1018.2118.3001.4449","title":null},"content":[{"type":"text","text":"字節跳動總結的設計模式 PDF 火了，完整版開放分享","attrs":{}}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https://blog.csdn.net/javachengzi/article/details/109729058?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522160665850419721940242314%2522%252C%2522scm%2522%253A%252220140713.130102334.pc%255Fall.%2522%257D&request_id=160665850419721940242314&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2~all~first_rank_v2~rank_v28-1-109729058.pc_first_rank_v2_rank_v28&utm_term=%E5%88%B7Github%E6%97%B6%E5%8F%91%E7%8E%B0%E4%BA%86%E4%B8%80%E6%9C%AC%E9%98%BF%E9%87%8C%E5%A4%A7%E7%A5%9E%E7%9A%84%E7%AE%97%E6%B3%95%E7%AC%94%E8%AE%B0%20%E6%A0%87%E6%98%9F70.5K&spm=1018.2118.3001.4449","title":null},"content":[{"type":"text","text":"刷Github時發現了一本阿里大神的算法筆記！標星70.5K","attrs":{}}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https://blog.csdn.net/javachengzi/article/details/108219072","title":null},"content":[{"type":"text","text":"如果能聽懂這個網約車實戰，哪怕接私活你都可以月入40K","attrs":{}}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https://blog.csdn.net/javachengzi/article/details/108219072","title":null},"content":[{"type":"text","text":"爲什麼阿里巴巴的程序員成長速度這麼快，看完他們的內部資料我懂了","attrs":{}}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https://www.bilibili.com/video/BV1f541157DX?spm_id_from=333.788.b_765f64657363.2","title":null},"content":[{"type":"text","text":"程序員達到50W年薪所需要具備的知識體系。","attrs":{}}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https://www.bilibili.com/video/BV1Ay4y167CV/","title":null},"content":[{"type":"text","text":"關於【暴力遞歸算法】你所不知道的思路","attrs":{}}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":1},"content":[{"type":"text","text":"看完三件事❤️","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"如果你覺得這篇內容對你還蠻有幫助，我想邀請你幫我三個小忙：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"點贊，轉發，有你們的 『點贊和評論』，纔是我創造的動力。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"關注公衆號 『 Java鬥帝 』，不定期分享原創知識。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"同時可以期待後續文章ing🚀","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}}]}