如果一個(gè)鍵過(guò)期了,那么它什么時(shí)候會(huì)被刪除呢?
這個(gè)問(wèn)題有三種可能的答案,它們分別代表了三種不同的刪除策略:
- 定時(shí)刪除:在設(shè)置鍵的過(guò)期時(shí)間的同時(shí),創(chuàng)建一個(gè)定時(shí)器( timer ). 讓定時(shí)器在鍵的過(guò)期時(shí)間來(lái)臨時(shí),立即執(zhí)行對(duì)鍵的刪除操作。
- 惰性刪除:放任鍵過(guò)期不管,但是每次從鍵空間中獲取鍵時(shí),都檢查取得的鍵是否過(guò)期,如果過(guò)期的話,就刪除該鍵;如果沒(méi)有過(guò)期,就返回該鍵。
- 定期刪除: 每隔一段時(shí)間,程序就對(duì)數(shù)據(jù)庫(kù)進(jìn)行一次檢查,刪除里面的過(guò)期鍵。至于要?jiǎng)h除多少過(guò)期鍵,以及要檢查多少個(gè)數(shù)據(jù)庫(kù), 則由算法決定。
在這三種策略中,第一種和第三種為主動(dòng)刪除策略, 而第二種則為被動(dòng)刪除策略。
前言
使用Redis時(shí)我們可以使用EXPIRE或EXPIREAT命令給key設(shè)置過(guò)期刪除時(shí)間,結(jié)構(gòu)體redisDb中的expires字典保存了所有key的過(guò)期時(shí)間,這個(gè)字典(dict)的key是一個(gè)指針,指向redis中的某個(gè)key對(duì)象,過(guò)期字典的value是一個(gè)保存過(guò)期時(shí)間的整數(shù)。
/* Redis database representation. There are multiple databases identified
* by integers from 0 (the default database) up to the max configured
* database. The database number is the 'id' field in the structure. */
typedef struct redisDb {
dict *dict; /* The keyspace for this DB */
dict *expires; /* 過(guò)期字典*/
dict *blocking_keys; /* Keys with clients waiting for data (BLPOP) */
dict *ready_keys; /* Blocked keys that received a PUSH */
dict *watched_keys; /* WATCHED keys for MULTI/EXEC CAS */
struct evictionPoolEntry *eviction_pool; /* Eviction pool of keys */
int id; /* Database ID */
long long avg_ttl; /* Average TTL, just for stats */
} redisDb;
設(shè)置過(guò)期時(shí)間
不論是EXPIRE,EXPIREAT,還是PEXPIRE,PEXPIREAT,底層的具體實(shí)現(xiàn)是一樣的。在Redis的key空間中找到要設(shè)置過(guò)期時(shí)間的這個(gè)key,然后將這個(gè)entry(key的指針,過(guò)期時(shí)間)加入到過(guò)期字典中。
void setExpire(redisDb *db, robj *key, long long when) {
dictEntry *kde, *de;
/* Reuse the sds from the main dict in the expire dict */
kde = dictFind(db->dict,key->ptr);
redisAssertWithInfo(NULL,key,kde != NULL);
de = dictReplaceRaw(db->expires,dictGetKey(kde));
dictSetSignedIntegerVal(de,when);
}
![](http://img.jbzj.com/file_images/article/201809/201899154459669.jpg?20188915456)
過(guò)期刪除策略
如果一個(gè)key過(guò)期了,何時(shí)會(huì)被刪除呢?在Redis中有兩種過(guò)期刪除策略:(1)惰性過(guò)期刪除;(2)定期刪除。接下來(lái)具體看看。
惰性過(guò)期刪除
Redis在執(zhí)行任何讀寫命令時(shí)都會(huì)先找到這個(gè)key,惰性刪除就作為一個(gè)切入點(diǎn)放在查找key之前,如果key過(guò)期了就刪除這個(gè)key。
![](http://img.jbzj.com/file_images/article/201809/201899154610647.jpg?201889154620)
robj *lookupKeyRead(redisDb *db, robj *key) {
robj *val;
expireIfNeeded(db,key); // 切入點(diǎn)
val = lookupKey(db,key);
if (val == NULL)
server.stat_keyspace_misses++;
else
server.stat_keyspace_hits++;
return val;
}
定期刪除
key的定期刪除會(huì)在Redis的周期性執(zhí)行任務(wù)(serverCron,默認(rèn)每100ms執(zhí)行一次)中進(jìn)行,而且是發(fā)生Redis的master節(jié)點(diǎn),因?yàn)閟lave節(jié)點(diǎn)會(huì)通過(guò)主節(jié)點(diǎn)的DEL命令同步過(guò)來(lái)達(dá)到刪除key的目的。
![](http://img.jbzj.com/file_images/article/201809/201899154714339.jpg?201889154725)
依次遍歷每個(gè)db(默認(rèn)配置數(shù)是16),針對(duì)每個(gè)db,每次循環(huán)隨機(jī)選擇20個(gè)(ACTIVE_EXPIRE_CYCLE_LOOKUPS_PER_LOOP)key判斷是否過(guò)期,如果一輪所選的key少于25%過(guò)期,則終止迭次,此外在迭代過(guò)程中如果超過(guò)了一定的時(shí)間限制則終止過(guò)期刪除這一過(guò)程。
for (j = 0; j dbs_per_call; j++) {
int expired;
redisDb *db = server.db+(current_db % server.dbnum);
/* Increment the DB now so we are sure if we run out of time
* in the current DB we'll restart from the next. This allows to
* distribute the time evenly across DBs. */
current_db++;
/* Continue to expire if at the end of the cycle more than 25%
* of the keys were expired. */
do {
unsigned long num, slots;
long long now, ttl_sum;
int ttl_samples;
/* 如果該db沒(méi)有設(shè)置過(guò)期key,則繼續(xù)看下個(gè)db*/
if ((num = dictSize(db->expires)) == 0) {
db->avg_ttl = 0;
break;
}
slots = dictSlots(db->expires);
now = mstime();
/* When there are less than 1% filled slots getting random
* keys is expensive, so stop here waiting for better times...
* The dictionary will be resized asap. */
if (num slots > DICT_HT_INITIAL_SIZE
(num*100/slots 1)) break;
/* The main collection cycle. Sample random keys among keys
* with an expire set, checking for expired ones. */
expired = 0;
ttl_sum = 0;
ttl_samples = 0;
if (num > ACTIVE_EXPIRE_CYCLE_LOOKUPS_PER_LOOP)
num = ACTIVE_EXPIRE_CYCLE_LOOKUPS_PER_LOOP;// 20
while (num--) {
dictEntry *de;
long long ttl;
if ((de = dictGetRandomKey(db->expires)) == NULL) break;
ttl = dictGetSignedIntegerVal(de)-now;
if (activeExpireCycleTryExpire(db,de,now)) expired++;
if (ttl > 0) {
/* We want the average TTL of keys yet not expired. */
ttl_sum += ttl;
ttl_samples++;
}
}
/* Update the average TTL stats for this database. */
if (ttl_samples) {
long long avg_ttl = ttl_sum/ttl_samples;
/* Do a simple running average with a few samples.
* We just use the current estimate with a weight of 2%
* and the previous estimate with a weight of 98%. */
if (db->avg_ttl == 0) db->avg_ttl = avg_ttl;
db->avg_ttl = (db->avg_ttl/50)*49 + (avg_ttl/50);
}
/* We can't block forever here even if there are many keys to
* expire. So after a given amount of milliseconds return to the
* caller waiting for the other active expire cycle. */
iteration++;
if ((iteration 0xf) == 0) { /* 每迭代16次檢查一次 */
long long elapsed = ustime()-start;
latencyAddSampleIfNeeded("expire-cycle",elapsed/1000);
if (elapsed > timelimit) timelimit_exit = 1;
}
// 超過(guò)時(shí)間限制則退出
if (timelimit_exit) return;
/* 在當(dāng)前db中,如果少于25%的key過(guò)期,則停止繼續(xù)刪除過(guò)期key */
} while (expired > ACTIVE_EXPIRE_CYCLE_LOOKUPS_PER_LOOP/4);
}
總結(jié)
惰性刪除:讀寫之前判斷key是否過(guò)期
定期刪除:定期抽樣key,判斷是否過(guò)期
好了,以上就是這篇文章的全部?jī)?nèi)容了,希望本文的內(nèi)容對(duì)大家的學(xué)習(xí)或者工作具有一定的參考學(xué)習(xí)價(jià)值,如果有疑問(wèn)大家可以留言交流,謝謝大家對(duì)腳本之家的支持。
您可能感興趣的文章:- Redis的LRU機(jī)制介紹
- Redis中的數(shù)據(jù)過(guò)期策略詳解
- 淺談redis的maxmemory設(shè)置以及淘汰策略
- 關(guān)于redis Key淘汰策略的實(shí)現(xiàn)方法
- Redis中LRU淘汰策略的深入分析