Nicksxs's Blog

What hurts more, the pain of hard work or the pain of regret?

0%

上一篇聊了mysql 的 innodb 引擎基于 read view 实现的 mvcc 和事务隔离级别,可能有些细心的小伙伴会发现一些问题,第一个是在 RC 级别下的事务提交后的可见性,这里涉及到了三个参数,m_low_limit_id,m_up_limit_id,m_ids,之前看到知乎的一篇写的非常不错的文章,但是就在这一点上似乎有点疑惑,这里基于源码和注释来解释下这个问题

1
2
3
4
5
6
7
8
9
10
11
/**
Opens a read view where exactly the transactions serialized before this
point in time are seen in the view.
@param id Creator transaction id */

void ReadView::prepare(trx_id_t id) {
ut_ad(mutex_own(&trx_sys->mutex));

m_creator_trx_id = id;

m_low_limit_no = m_low_limit_id = m_up_limit_id = trx_sys->max_trx_id;

m_low_limit_id赋的值是trx_sys->max_trx_id,代表的是当前系统最小的未分配的事务 id,所以呢,举个例子,当前有三个活跃事务,事务 id 分别是 100,200,300,而 m_up_limit_id = 100, m_low_limit_id = 301,当事务 id 是 200 的提交之后,它的更新就是可以被 100 和 300 看到,而不是说 m_ids 里没了 200,并且 200 比 100 大就应该不可见了

幻读

还有一个问题是幻读的问题,这貌似也是个高频面试题,啥意思呢,或者说跟它最常拿来比较的脏读,脏读是指读到了别的事务未提交的数据,因为未提交,严格意义上来讲,不一定是会被最后落到库里,可能会回滚,也就是在 read uncommitted 级别下会出现的问题,但是幻读不太一样,幻读是指两次查询的结果数量不一样,比如我查了第一次是 select * from table1 where id < 10 for update,查出来了一条结果 id 是 5,然后再查一下发现出来了一条 id 是 5,一条 id 是 7,那是不是有点尴尬了,其实呢这个点我觉得脏读跟幻读也比较是从原理层面来命名,如果第一次接触的同学发觉有点不理解也比较正常,因为从逻辑上讲总之都是数据不符合预期,但是基于源码层面其实是不同的情况,幻读是在原先的 read view 无法完全解决的,怎么解决呢,简单的来说就是锁咯,我们知道innodb 是基于 record lock 行锁的,但是貌似没有办法解决这种问题,那么 innodb 就引入了 gap lock 间隙锁,比如上面说的情况下,id 小于 10 的情况下,是都应该锁住的,gap lock 其实是基于索引结构来锁的,因为索引树除了树形结构之外,还有一个next record 的指针,gap lock 也是基于这个来锁的
看一下 mysql 的文档

SELECT … FOR UPDATE sets an exclusive next-key lock on every record the search encounters. However, only an index record lock is required for statements that lock rows using a unique index to search for a unique row.

对于一个 for update 查询,在 RR 级别下,会设置一个 next-key lock在每一条被查询到的记录上,next-lock 又是啥呢,其实就是 gap 锁和 record 锁的结合体,比如我在数据库里有 id 是 1,3,5,7,10,对于上面那条查询,查出来的结果就是 1,3,5,7,那么按照文档里描述的,对于这几条记录都会加上next-key lock,也就是(-∞, 1], (1, 3], (3, 5], (5, 7], (7, 10) 这些区间和记录会被锁起来,不让插入,再唠叨一下呢,就是其实如果是只读的事务,光 read view 一致性读就够了,如果是有写操作的呢,就需要锁了。

很久以前,有位面试官问到,你知道 mysql 的事务隔离级别吗,“额 O__O …,不太清楚”,完了之后我就去网上找相关的文章,找到了这篇MySQL 四种事务隔离级的说明, 文章写得特别好,看了这个就懂了各个事务隔离级别都是啥,不过看了这个之后多思考一下的话还是会发现问题,这么神奇的事务隔离级别是怎么实现的呢

其中 innodb 的事务隔离用到了标题里说到的 mvcc,Multiversion concurrency control, 直译过来就是多版本并发控制,先不讲这个究竟是个啥,考虑下如果纯猜测,这个事务隔离级别应该会是怎么样实现呢,愚钝的我想了下,可以在事务开始的时候拷贝一个表,这个可以支持 RR 级别,RC 级别就不支持了,而且要是个非常大的表,想想就不可行

腆着脸说虽然这个不可行,但是思路是对的,具体实行起来需要做一系列(肥肠多)的改动,首先根据我的理解,其实这个拷贝一个表是变成拷贝一条记录,但是如果有多个事务,那就得拷贝多次,这个问题其实可以借助版本管理系统来解释,在用版本管理系统,git 之类的之前,很原始的可能是开发完一个功能后,就打个压缩包用时间等信息命名,然后如果后面要找回这个就直接用这个压缩包的就行了,后来有了 svn,git 中心式和分布式的版本管理系统,它的一个特点是粒度可以控制到文件和代码行级别,对应的我们的 mysql 事务是不是也可以从一开始预想的表级别细化到行的级别,可能之前很多人都了解过,数据库的一行记录除了我们用户自定义的字段,还有一些额外的字段,去源码data0type.h里捞一下

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
/* Precise data types for system columns and the length of those columns;
NOTE: the values must run from 0 up in the order given! All codes must
be less than 256 */
#define DATA_ROW_ID 0 /* row id: a 48-bit integer */
#define DATA_ROW_ID_LEN 6 /* stored length for row id */

/** Transaction id: 6 bytes */
constexpr size_t DATA_TRX_ID = 1;

/** Transaction ID type size in bytes. */
constexpr size_t DATA_TRX_ID_LEN = 6;

/** Rollback data pointer: 7 bytes */
constexpr size_t DATA_ROLL_PTR = 2;

/** Rollback data pointer type size in bytes. */
constexpr size_t DATA_ROLL_PTR_LEN = 7;

一个是 DATA_ROW_ID,这个是在数据没指定主键的时候会生成一个隐藏的,如果用户有指定主键就是主键了

一个是 DATA_TRX_ID,这个表示这条记录的事务 ID

还有一个是 DATA_ROLL_PTR 指向回滚段的指针

指向的回滚段其实就是我们常说的 undo log,这里面的具体结构就是个链表,在 mvcc 里会使用到这个,还有就是这个 DATA_TRX_ID,每条记录都记录了这个事务 ID,表示的是这条记录的当前值是被哪个事务修改的,下面就扯回事务了,我们知道 Read Uncommitted, 其实用不到隔离,直接读取当前值就好了,到了 Read Committed 级别,我们要让事务读取到提交过的值,mysql 使用了一个叫 read view 的玩意,它里面有这些值是我们需要注意的,

m_low_limit_id, 这个是 read view 创建时最大的活跃事务 id

m_up_limit_id, 这个是 read view 创建时最小的活跃事务 id

m_ids, 这个是 read view 创建时所有的活跃事务 id 数组

m_creator_trx_id 这个是当前记录的创建事务 id

判断事务的可见性主要的逻辑是这样,

  1. 当记录的事务 id 小于最小活跃事务 id,说明是可见的,
  2. 如果记录的事务 id 等于当前事务 id,说明是自己的更改,可见
  3. 如果记录的事务 id 大于最大的活跃事务 id, 不可见
  4. 如果记录的事务 id 介于 m_low_limit_idm_up_limit_id 之间,则要判断它是否在 m_ids 中,如果在,不可见,如果不在,表示已提交,可见
    具体的代码捞一下看看
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    22
    23
    24
    25
    /** Check whether the changes by id are visible.
    @param[in] id transaction id to check against the view
    @param[in] name table name
    @return whether the view sees the modifications of id. */
    bool changes_visible(trx_id_t id, const table_name_t &name) const
    MY_ATTRIBUTE((warn_unused_result)) {
    ut_ad(id > 0);

    if (id < m_up_limit_id || id == m_creator_trx_id) {
    return (true);
    }

    check_trx_id_sanity(id, name);

    if (id >= m_low_limit_id) {
    return (false);

    } else if (m_ids.empty()) {
    return (true);
    }

    const ids_t::value_type *p = m_ids.data();

    return (!std::binary_search(p, p + m_ids.size(), id));
    }
    剩下来一点是啥呢,就是 Read CommittedRepeated Read 也不一样,那前面说的 read view 都能支持吗,又是怎么支持呢,假如这个 read view 是在事务一开始就创建,那好像能支持的只是 RR 事务隔离级别,其实呢,这是通过创建 read view的时机,对于 RR 级别,就是在事务的第一个 select 语句是创建,对于 RC 级别,是在每个 select 语句执行前都是创建一次,那样就可以保证能读到所有已提交的数据

LRU

说完了过期策略再说下淘汰策略,redis 使用的策略是近似的 lru 策略,为什么是近似的呢,先来看下什么是 lru,看下 wiki 的介绍
,图中一共有四个槽的存储空间,依次访问顺序是 A B C D E D F,
当第一次访问 D 时刚好占满了坑,并且值是 4,这个值越小代表越先被淘汰,当 E 进来时,看了下已经存在的四个里 A 是最小的,代表是最早存在并且最早被访问的,那就先淘汰它了,E 占领了 A 的位置,并设置值为 4,然后又访问 D 了,D 已经存在了,不过又被访问到了,得更新值为 5,然后是 F 进来了,这时 B 是最老的且最近未被访问,所以就淘汰它了。以上是一个 lru 的简要说明,但是 redis 没有严格按照这个去执行,理由跟前面过期策略一致,最严格的过期策略应该是每个 key 都有对应的定时器,当超时时马上就能清除,但是问题是这样的cpu 消耗太大,所换来的内存效率不太值得,淘汰策略也是这样,类似于上图,要维护所有 key 的一个有序 lru 值,并且遍历将最小的淘汰,redis 采用的是抽样的形式,最初的实现方式是随机从 dict 抽取 5 个 key,淘汰一个 lru 最小的,这样子勉强能达到淘汰的目的,但是效果不是特别好,后面在 redis 3.0开始,将随机抽取改成了维护一个 pool,pool 的大小默认是 16,每次放入的都是按lru 值有序排列好,每一次放入的必须是 lru小于 pool 中最小的 lru 才允许放入,直到放满,后面再有新的就会将大的踢出。
redis 针对这个策略的改进做了一个实验,这里借用下图

首先背景是这图中的所有点都对应一个 redis 的 key,灰色部分加入后被顺序访问过一遍,然后又加入了绿色部分,那么按照理论的 lru 算法,应该是图左上中,浅灰色部分全都被淘汰,那么对比来看看图右上,左下和右下,左下表示 2.8 版本就是随机抽样 5 个 key,淘汰其中 lru 最小的一个,发现是灰色和浅灰色的都有被淘汰的,右下的 3.0 版本抽样数量不变的情况下,稍好一些,当 3.0 版本的抽样数量调整成 10 后,已经较为接近理论上的 lru 策略了,通过代码来简要分析下

1
2
3
4
5
6
7
8
9
typedef struct redisObject {
unsigned type:4;
unsigned encoding:4;
unsigned lru:LRU_BITS; /* LRU time (relative to global lru_clock) or
* LFU data (least significant 8 bits frequency
* and most significant 16 bits access time). */
int refcount;
void *ptr;
} robj;

对于 lru 策略来说,lru 字段记录的就是redisObj 的LRU time,
redis 在访问数据时,都会调用lookupKey方法

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
/* Low level key lookup API, not actually called directly from commands
* implementations that should instead rely on lookupKeyRead(),
* lookupKeyWrite() and lookupKeyReadWithFlags(). */
robj *lookupKey(redisDb *db, robj *key, int flags) {
dictEntry *de = dictFind(db->dict,key->ptr);
if (de) {
robj *val = dictGetVal(de);

/* Update the access time for the ageing algorithm.
* Don't do it if we have a saving child, as this will trigger
* a copy on write madness. */
if (!hasActiveChildProcess() && !(flags & LOOKUP_NOTOUCH)){
if (server.maxmemory_policy & MAXMEMORY_FLAG_LFU) {
// 这个是后面一节的内容
updateLFU(val);
} else {
// 对于这个分支,访问时就会去更新 lru 值
val->lru = LRU_CLOCK();
}
}
return val;
} else {
return NULL;
}
}
/* This function is used to obtain the current LRU clock.
* If the current resolution is lower than the frequency we refresh the
* LRU clock (as it should be in production servers) we return the
* precomputed value, otherwise we need to resort to a system call. */
unsigned int LRU_CLOCK(void) {
unsigned int lruclock;
if (1000/server.hz <= LRU_CLOCK_RESOLUTION) {
// 如果服务器的频率server.hz大于 1 时就是用系统预设的 lruclock
lruclock = server.lruclock;
} else {
lruclock = getLRUClock();
}
return lruclock;
}
/* Return the LRU clock, based on the clock resolution. This is a time
* in a reduced-bits format that can be used to set and check the
* object->lru field of redisObject structures. */
unsigned int getLRUClock(void) {
return (mstime()/LRU_CLOCK_RESOLUTION) & LRU_CLOCK_MAX;
}

redis 处理命令是在这里processCommand

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
/* If this function gets called we already read a whole
* command, arguments are in the client argv/argc fields.
* processCommand() execute the command or prepare the
* server for a bulk read from the client.
*
* If C_OK is returned the client is still alive and valid and
* other operations can be performed by the caller. Otherwise
* if C_ERR is returned the client was destroyed (i.e. after QUIT). */
int processCommand(client *c) {
moduleCallCommandFilters(c);



/* Handle the maxmemory directive.
*
* Note that we do not want to reclaim memory if we are here re-entering
* the event loop since there is a busy Lua script running in timeout
* condition, to avoid mixing the propagation of scripts with the
* propagation of DELs due to eviction. */
if (server.maxmemory && !server.lua_timedout) {
int out_of_memory = freeMemoryIfNeededAndSafe() == C_ERR;
/* freeMemoryIfNeeded may flush slave output buffers. This may result
* into a slave, that may be the active client, to be freed. */
if (server.current_client == NULL) return C_ERR;

/* It was impossible to free enough memory, and the command the client
* is trying to execute is denied during OOM conditions or the client
* is in MULTI/EXEC context? Error. */
if (out_of_memory &&
(c->cmd->flags & CMD_DENYOOM ||
(c->flags & CLIENT_MULTI &&
c->cmd->proc != execCommand &&
c->cmd->proc != discardCommand)))
{
flagTransaction(c);
addReply(c, shared.oomerr);
return C_OK;
}
}
}

这里只摘了部分,当需要清理内存时就会调用, 然后调用了freeMemoryIfNeededAndSafe

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
/* This is a wrapper for freeMemoryIfNeeded() that only really calls the
* function if right now there are the conditions to do so safely:
*
* - There must be no script in timeout condition.
* - Nor we are loading data right now.
*
*/
int freeMemoryIfNeededAndSafe(void) {
if (server.lua_timedout || server.loading) return C_OK;
return freeMemoryIfNeeded();
}
/* This function is periodically called to see if there is memory to free
* according to the current "maxmemory" settings. In case we are over the
* memory limit, the function will try to free some memory to return back
* under the limit.
*
* The function returns C_OK if we are under the memory limit or if we
* were over the limit, but the attempt to free memory was successful.
* Otehrwise if we are over the memory limit, but not enough memory
* was freed to return back under the limit, the function returns C_ERR. */
int freeMemoryIfNeeded(void) {
int keys_freed = 0;
/* By default replicas should ignore maxmemory
* and just be masters exact copies. */
if (server.masterhost && server.repl_slave_ignore_maxmemory) return C_OK;

size_t mem_reported, mem_tofree, mem_freed;
mstime_t latency, eviction_latency;
long long delta;
int slaves = listLength(server.slaves);

/* When clients are paused the dataset should be static not just from the
* POV of clients not being able to write, but also from the POV of
* expires and evictions of keys not being performed. */
if (clientsArePaused()) return C_OK;
if (getMaxmemoryState(&mem_reported,NULL,&mem_tofree,NULL) == C_OK)
return C_OK;

mem_freed = 0;

if (server.maxmemory_policy == MAXMEMORY_NO_EVICTION)
goto cant_free; /* We need to free memory, but policy forbids. */

latencyStartMonitor(latency);
while (mem_freed < mem_tofree) {
int j, k, i;
static unsigned int next_db = 0;
sds bestkey = NULL;
int bestdbid;
redisDb *db;
dict *dict;
dictEntry *de;

if (server.maxmemory_policy & (MAXMEMORY_FLAG_LRU|MAXMEMORY_FLAG_LFU) ||
server.maxmemory_policy == MAXMEMORY_VOLATILE_TTL)
{
struct evictionPoolEntry *pool = EvictionPoolLRU;

while(bestkey == NULL) {
unsigned long total_keys = 0, keys;

/* We don't want to make local-db choices when expiring keys,
* so to start populate the eviction pool sampling keys from
* every DB. */
for (i = 0; i < server.dbnum; i++) {
db = server.db+i;
dict = (server.maxmemory_policy & MAXMEMORY_FLAG_ALLKEYS) ?
db->dict : db->expires;
if ((keys = dictSize(dict)) != 0) {
evictionPoolPopulate(i, dict, db->dict, pool);
total_keys += keys;
}
}
if (!total_keys) break; /* No keys to evict. */

/* Go backward from best to worst element to evict. */
for (k = EVPOOL_SIZE-1; k >= 0; k--) {
if (pool[k].key == NULL) continue;
bestdbid = pool[k].dbid;

if (server.maxmemory_policy & MAXMEMORY_FLAG_ALLKEYS) {
de = dictFind(server.db[pool[k].dbid].dict,
pool[k].key);
} else {
de = dictFind(server.db[pool[k].dbid].expires,
pool[k].key);
}

/* Remove the entry from the pool. */
if (pool[k].key != pool[k].cached)
sdsfree(pool[k].key);
pool[k].key = NULL;
pool[k].idle = 0;

/* If the key exists, is our pick. Otherwise it is
* a ghost and we need to try the next element. */
if (de) {
bestkey = dictGetKey(de);
break;
} else {
/* Ghost... Iterate again. */
}
}
}
}

/* volatile-random and allkeys-random policy */
else if (server.maxmemory_policy == MAXMEMORY_ALLKEYS_RANDOM ||
server.maxmemory_policy == MAXMEMORY_VOLATILE_RANDOM)
{
/* When evicting a random key, we try to evict a key for
* each DB, so we use the static 'next_db' variable to
* incrementally visit all DBs. */
for (i = 0; i < server.dbnum; i++) {
j = (++next_db) % server.dbnum;
db = server.db+j;
dict = (server.maxmemory_policy == MAXMEMORY_ALLKEYS_RANDOM) ?
db->dict : db->expires;
if (dictSize(dict) != 0) {
de = dictGetRandomKey(dict);
bestkey = dictGetKey(de);
bestdbid = j;
break;
}
}
}

/* Finally remove the selected key. */
if (bestkey) {
db = server.db+bestdbid;
robj *keyobj = createStringObject(bestkey,sdslen(bestkey));
propagateExpire(db,keyobj,server.lazyfree_lazy_eviction);
/* We compute the amount of memory freed by db*Delete() alone.
* It is possible that actually the memory needed to propagate
* the DEL in AOF and replication link is greater than the one
* we are freeing removing the key, but we can't account for
* that otherwise we would never exit the loop.
*
* AOF and Output buffer memory will be freed eventually so
* we only care about memory used by the key space. */
delta = (long long) zmalloc_used_memory();
latencyStartMonitor(eviction_latency);
if (server.lazyfree_lazy_eviction)
dbAsyncDelete(db,keyobj);
else
dbSyncDelete(db,keyobj);
latencyEndMonitor(eviction_latency);
latencyAddSampleIfNeeded("eviction-del",eviction_latency);
latencyRemoveNestedEvent(latency,eviction_latency);
delta -= (long long) zmalloc_used_memory();
mem_freed += delta;
server.stat_evictedkeys++;
notifyKeyspaceEvent(NOTIFY_EVICTED, "evicted",
keyobj, db->id);
decrRefCount(keyobj);
keys_freed++;

/* When the memory to free starts to be big enough, we may
* start spending so much time here that is impossible to
* deliver data to the slaves fast enough, so we force the
* transmission here inside the loop. */
if (slaves) flushSlavesOutputBuffers();

/* Normally our stop condition is the ability to release
* a fixed, pre-computed amount of memory. However when we
* are deleting objects in another thread, it's better to
* check, from time to time, if we already reached our target
* memory, since the "mem_freed" amount is computed only
* across the dbAsyncDelete() call, while the thread can
* release the memory all the time. */
if (server.lazyfree_lazy_eviction && !(keys_freed % 16)) {
if (getMaxmemoryState(NULL,NULL,NULL,NULL) == C_OK) {
/* Let's satisfy our stop condition. */
mem_freed = mem_tofree;
}
}
} else {
latencyEndMonitor(latency);
latencyAddSampleIfNeeded("eviction-cycle",latency);
goto cant_free; /* nothing to free... */
}
}
latencyEndMonitor(latency);
latencyAddSampleIfNeeded("eviction-cycle",latency);
return C_OK;

cant_free:
/* We are here if we are not able to reclaim memory. There is only one
* last thing we can try: check if the lazyfree thread has jobs in queue
* and wait... */
while(bioPendingJobsOfType(BIO_LAZY_FREE)) {
if (((mem_reported - zmalloc_used_memory()) + mem_freed) >= mem_tofree)
break;
usleep(1000);
}
return C_ERR;
}

这里就是根据具体策略去淘汰 key,首先是要往 pool 更新 key,更新key 的方法是evictionPoolPopulate

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
void evictionPoolPopulate(int dbid, dict *sampledict, dict *keydict, struct evictionPoolEntry *pool) {
int j, k, count;
dictEntry *samples[server.maxmemory_samples];

count = dictGetSomeKeys(sampledict,samples,server.maxmemory_samples);
for (j = 0; j < count; j++) {
unsigned long long idle;
sds key;
robj *o;
dictEntry *de;

de = samples[j];
key = dictGetKey(de);

/* If the dictionary we are sampling from is not the main
* dictionary (but the expires one) we need to lookup the key
* again in the key dictionary to obtain the value object. */
if (server.maxmemory_policy != MAXMEMORY_VOLATILE_TTL) {
if (sampledict != keydict) de = dictFind(keydict, key);
o = dictGetVal(de);
}

/* Calculate the idle time according to the policy. This is called
* idle just because the code initially handled LRU, but is in fact
* just a score where an higher score means better candidate. */
if (server.maxmemory_policy & MAXMEMORY_FLAG_LRU) {
idle = estimateObjectIdleTime(o);
} else if (server.maxmemory_policy & MAXMEMORY_FLAG_LFU) {
/* When we use an LRU policy, we sort the keys by idle time
* so that we expire keys starting from greater idle time.
* However when the policy is an LFU one, we have a frequency
* estimation, and we want to evict keys with lower frequency
* first. So inside the pool we put objects using the inverted
* frequency subtracting the actual frequency to the maximum
* frequency of 255. */
idle = 255-LFUDecrAndReturn(o);
} else if (server.maxmemory_policy == MAXMEMORY_VOLATILE_TTL) {
/* In this case the sooner the expire the better. */
idle = ULLONG_MAX - (long)dictGetVal(de);
} else {
serverPanic("Unknown eviction policy in evictionPoolPopulate()");
}

/* Insert the element inside the pool.
* First, find the first empty bucket or the first populated
* bucket that has an idle time smaller than our idle time. */
k = 0;
while (k < EVPOOL_SIZE &&
pool[k].key &&
pool[k].idle < idle) k++;
if (k == 0 && pool[EVPOOL_SIZE-1].key != NULL) {
/* Can't insert if the element is < the worst element we have
* and there are no empty buckets. */
continue;
} else if (k < EVPOOL_SIZE && pool[k].key == NULL) {
/* Inserting into empty position. No setup needed before insert. */
} else {
/* Inserting in the middle. Now k points to the first element
* greater than the element to insert. */
if (pool[EVPOOL_SIZE-1].key == NULL) {
/* Free space on the right? Insert at k shifting
* all the elements from k to end to the right. */

/* Save SDS before overwriting. */
sds cached = pool[EVPOOL_SIZE-1].cached;
memmove(pool+k+1,pool+k,
sizeof(pool[0])*(EVPOOL_SIZE-k-1));
pool[k].cached = cached;
} else {
/* No free space on right? Insert at k-1 */
k--;
/* Shift all elements on the left of k (included) to the
* left, so we discard the element with smaller idle time. */
sds cached = pool[0].cached; /* Save SDS before overwriting. */
if (pool[0].key != pool[0].cached) sdsfree(pool[0].key);
memmove(pool,pool+1,sizeof(pool[0])*k);
pool[k].cached = cached;
}
}

/* Try to reuse the cached SDS string allocated in the pool entry,
* because allocating and deallocating this object is costly
* (according to the profiler, not my fantasy. Remember:
* premature optimizbla bla bla bla. */
int klen = sdslen(key);
if (klen > EVPOOL_CACHED_SDS_SIZE) {
pool[k].key = sdsdup(key);
} else {
memcpy(pool[k].cached,key,klen+1);
sdssetlen(pool[k].cached,klen);
pool[k].key = pool[k].cached;
}
pool[k].idle = idle;
pool[k].dbid = dbid;
}
}

Redis随机选择maxmemory_samples数量的key,然后计算这些key的空闲时间idle time,当满足条件时(比pool中的某些键的空闲时间还大)就可以进poolpool更新之后,就淘汰pool中空闲时间最大的键。

estimateObjectIdleTime用来计算Redis对象的空闲时间:

1
2
3
4
5
6
7
8
9
10
11
/* Given an object returns the min number of milliseconds the object was never
* requested, using an approximated LRU algorithm. */
unsigned long long estimateObjectIdleTime(robj *o) {
unsigned long long lruclock = LRU_CLOCK();
if (lruclock >= o->lru) {
return (lruclock - o->lru) * LRU_CLOCK_RESOLUTION;
} else {
return (lruclock + (LRU_CLOCK_MAX - o->lru)) *
LRU_CLOCK_RESOLUTION;
}
}

空闲时间第一种是 lurclock 大于对象的 lru,那么就是减一下乘以精度,因为 lruclock 有可能是已经预生成的,所以会可能走下面这个

LFU

上面介绍了LRU 的算法,但是考虑一种场景

1
2
3
4
~~~~~A~~~~~A~~~~~A~~~~A~~~~~A~~~~~A~~|
~~B~~B~~B~~B~~B~~B~~B~~B~~B~~B~~B~~B~|
~~~~~~~~~~C~~~~~~~~~C~~~~~~~~~C~~~~~~|
~~~~~D~~~~~~~~~~D~~~~~~~~~D~~~~~~~~~D|

可以发现,当采用 lru 的淘汰策略的时候,D 是最新的,会被认为是最值得保留的,但是事实上还不如 A 跟 B,然后 antirez 大神就想到了LFU (Least Frequently Used) 这个算法, 显然对于上面的四个 key 的访问频率,保留优先级应该是 B > A > C = D
那要怎么来实现这个 LFU 算法呢,其实像LRU,理想的情况就是维护个链表,把最新访问的放到头上去,但是这个会影响访问速度,注意到前面代码的应该可以看到,redisObject 的 lru 字段其实是两用的,当策略是 LFU 时,这个字段就另作他用了,它的 24 位长度被分成两部分

1
2
3
4
      16 bits      8 bits
+----------------+--------+
+ Last decr time | LOG_C |
+----------------+--------+

前16位字段是最后一次递减时间,因此Redis知道 上一次计数器递减,后8位是 计数器 counter。
LFU 的主体策略就是当这个 key 被访问的次数越多频率越高他就越容易被保留下来,并且是最近被访问的频率越高。这其实有两个事情要做,一个是在访问的时候增加计数值,在一定长时间不访问时进行衰减,所以这里用了两个值,前 16 位记录上一次衰减的时间,后 8 位记录具体的计数值。
Redis4.0之后为maxmemory_policy淘汰策略添加了两个LFU模式:

volatile-lfu:对有过期时间的key采用LFU淘汰策略
allkeys-lfu:对全部key采用LFU淘汰策略
还有2个配置可以调整LFU算法:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
lfu-log-factor 10
lfu-decay-time 1
```
`lfu-log-factor` 可以调整计数器counter的增长速度,lfu-log-factor越大,counter增长的越慢。

`lfu-decay-time`是一个以分钟为单位的数值,可以调整counter的减少速度
这里有个问题是 8 位大小够计么,访问一次加 1 的话的确不够,不过大神就是大神,才不会这么简单的加一。往下看代码
```C
/* Low level key lookup API, not actually called directly from commands
* implementations that should instead rely on lookupKeyRead(),
* lookupKeyWrite() and lookupKeyReadWithFlags(). */
robj *lookupKey(redisDb *db, robj *key, int flags) {
dictEntry *de = dictFind(db->dict,key->ptr);
if (de) {
robj *val = dictGetVal(de);

/* Update the access time for the ageing algorithm.
* Don't do it if we have a saving child, as this will trigger
* a copy on write madness. */
if (!hasActiveChildProcess() && !(flags & LOOKUP_NOTOUCH)){
if (server.maxmemory_policy & MAXMEMORY_FLAG_LFU) {
// 当淘汰策略是 LFU 时,就会调用这个updateLFU
updateLFU(val);
} else {
val->lru = LRU_CLOCK();
}
}
return val;
} else {
return NULL;
}
}

updateLFU 这个其实个入口,调用了两个重要的方法

1
2
3
4
5
6
7
8
/* Update LFU when an object is accessed.
* Firstly, decrement the counter if the decrement time is reached.
* Then logarithmically increment the counter, and update the access time. */
void updateLFU(robj *val) {
unsigned long counter = LFUDecrAndReturn(val);
counter = LFULogIncr(counter);
val->lru = (LFUGetTimeInMinutes()<<8) | counter;
}

首先来看看LFUDecrAndReturn,这个方法的作用是根据上一次衰减时间和系统配置的 lfu-decay-time 参数来确定需要将 counter 减去多少

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
/* If the object decrement time is reached decrement the LFU counter but
* do not update LFU fields of the object, we update the access time
* and counter in an explicit way when the object is really accessed.
* And we will times halve the counter according to the times of
* elapsed time than server.lfu_decay_time.
* Return the object frequency counter.
*
* This function is used in order to scan the dataset for the best object
* to fit: as we check for the candidate, we incrementally decrement the
* counter of the scanned objects if needed. */
unsigned long LFUDecrAndReturn(robj *o) {
// 右移 8 位,拿到上次衰减时间
unsigned long ldt = o->lru >> 8;
// 对 255 做与操作,拿到 counter 值
unsigned long counter = o->lru & 255;
// 根据lfu_decay_time来算出过了多少个衰减周期
unsigned long num_periods = server.lfu_decay_time ? LFUTimeElapsed(ldt) / server.lfu_decay_time : 0;
if (num_periods)
counter = (num_periods > counter) ? 0 : counter - num_periods;
return counter;
}

然后是加,调用了LFULogIncr

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
/* Logarithmically increment a counter. The greater is the current counter value
* the less likely is that it gets really implemented. Saturate it at 255. */
uint8_t LFULogIncr(uint8_t counter) {
// 最大值就是 255,到顶了就不加了
if (counter == 255) return 255;
// 生成个随机小数
double r = (double)rand()/RAND_MAX;
// 减去个基础值,LFU_INIT_VAL = 5,防止刚进来就被逐出
double baseval = counter - LFU_INIT_VAL;
// 如果是小于 0,
if (baseval < 0) baseval = 0;
// 如果 baseval 是 0,那么 p 就是 1了,后面 counter 直接加一,如果不是的话,得看系统参数lfu_log_factor,这个越大,除出来的 p 越小,那么 counter++的可能性也越小,这样子就把前面的疑问给解决了,不是直接+1 的
double p = 1.0/(baseval*server.lfu_log_factor+1);
if (r < p) counter++;
return counter;
}

大概的变化速度可以参考

1
2
3
4
5
6
7
8
9
10
11
+--------+------------+------------+------------+------------+------------+
| factor | 100 hits | 1000 hits | 100K hits | 1M hits | 10M hits |
+--------+------------+------------+------------+------------+------------+
| 0 | 104 | 255 | 255 | 255 | 255 |
+--------+------------+------------+------------+------------+------------+
| 1 | 18 | 49 | 255 | 255 | 255 |
+--------+------------+------------+------------+------------+------------+
| 10 | 10 | 18 | 142 | 255 | 255 |
+--------+------------+------------+------------+------------+------------+
| 100 | 8 | 11 | 49 | 143 | 255 |
+--------+------------+------------+------------+------------+------------+

简而言之就是 lfu_log_factor 越大变化的越慢

总结

总结一下,redis 实现了近似的 lru 淘汰策略,通过增加了淘汰 key 的池子(pool),并且增大每次抽样的 key 的数量来将淘汰效果更进一步地接近于 lru,这是 lru 策略,但是对于前面举的一个例子,其实 lru 并不能保证 key 的淘汰就如我们预期,所以在后期又引入了 lfu 的策略,lfu的策略比较巧妙,复用了 redis 对象的 lru 字段,并且使用了factor 参数来控制计数器递增的速度,防止 8 位的计数器太早溢出。

这一篇不再是数据结构介绍了,大致的数据结构基本都介绍了,这一篇主要是查漏补缺,或者说讲一些重要且基本的概念,也可能是经常被忽略的,很多讲 redis 的系列文章可能都会忽略,学习 redis 的时候也会,因为觉得源码学习就是讲主要的数据结构和“算法”学习了就好了。
redis 的主要应用就是拿来作为高性能的缓存,那么缓存一般有些啥需要注意的,首先是访问速度,如果取得跟数据库一样快,那就没什么存在的意义,第二个是缓存的字面意思,我只是为了让数据读取快一些,通常大部分的场景这个是需要更新过期的,这里就把我要讲的第一点引出来了(真累,

redis过期策略

redis 是如何过期缓存的,可以猜测下,最无脑的就是每个设置了过期时间的 key 都设个定时器,过期了就删除,这种显然消耗太大,清理地最及时,还有的就是 redis 正在采用的懒汉清理策略和定期清理
懒汉策略就是在使用的时候去检查缓存是否过期,比如 get 操作时,先判断下这个 key 是否已经过期了,如果过期了就删掉,并且返回空,如果没过期则正常返回
主要代码是

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
/* This function is called when we are going to perform some operation
* in a given key, but such key may be already logically expired even if
* it still exists in the database. The main way this function is called
* is via lookupKey*() family of functions.
*
* The behavior of the function depends on the replication role of the
* instance, because slave instances do not expire keys, they wait
* for DELs from the master for consistency matters. However even
* slaves will try to have a coherent return value for the function,
* so that read commands executed in the slave side will be able to
* behave like if the key is expired even if still present (because the
* master has yet to propagate the DEL).
*
* In masters as a side effect of finding a key which is expired, such
* key will be evicted from the database. Also this may trigger the
* propagation of a DEL/UNLINK command in AOF / replication stream.
*
* The return value of the function is 0 if the key is still valid,
* otherwise the function returns 1 if the key is expired. */
int expireIfNeeded(redisDb *db, robj *key) {
if (!keyIsExpired(db,key)) return 0;

/* If we are running in the context of a slave, instead of
* evicting the expired key from the database, we return ASAP:
* the slave key expiration is controlled by the master that will
* send us synthesized DEL operations for expired keys.
*
* Still we try to return the right information to the caller,
* that is, 0 if we think the key should be still valid, 1 if
* we think the key is expired at this time. */
if (server.masterhost != NULL) return 1;

/* Delete the key */
server.stat_expiredkeys++;
propagateExpire(db,key,server.lazyfree_lazy_expire);
notifyKeyspaceEvent(NOTIFY_EXPIRED,
"expired",key,db->id);
return server.lazyfree_lazy_expire ? dbAsyncDelete(db,key) :
dbSyncDelete(db,key);
}

/* Check if the key is expired. */
int keyIsExpired(redisDb *db, robj *key) {
mstime_t when = getExpire(db,key);
mstime_t now;

if (when < 0) return 0; /* No expire for this key */

/* Don't expire anything while loading. It will be done later. */
if (server.loading) return 0;

/* If we are in the context of a Lua script, we pretend that time is
* blocked to when the Lua script started. This way a key can expire
* only the first time it is accessed and not in the middle of the
* script execution, making propagation to slaves / AOF consistent.
* See issue #1525 on Github for more information. */
if (server.lua_caller) {
now = server.lua_time_start;
}
/* If we are in the middle of a command execution, we still want to use
* a reference time that does not change: in that case we just use the
* cached time, that we update before each call in the call() function.
* This way we avoid that commands such as RPOPLPUSH or similar, that
* may re-open the same key multiple times, can invalidate an already
* open object in a next call, if the next call will see the key expired,
* while the first did not. */
else if (server.fixed_time_expire > 0) {
now = server.mstime;
}
/* For the other cases, we want to use the most fresh time we have. */
else {
now = mstime();
}

/* The key expired if the current (virtual or real) time is greater
* than the expire time of the key. */
return now > when;
}
/* Return the expire time of the specified key, or -1 if no expire
* is associated with this key (i.e. the key is non volatile) */
long long getExpire(redisDb *db, robj *key) {
dictEntry *de;

/* No expire? return ASAP */
if (dictSize(db->expires) == 0 ||
(de = dictFind(db->expires,key->ptr)) == NULL) return -1;

/* The entry was found in the expire dict, this means it should also
* be present in the main dict (safety check). */
serverAssertWithInfo(NULL,key,dictFind(db->dict,key->ptr) != NULL);
return dictGetSignedIntegerVal(de);
}

这里有几点要注意的,第一是当惰性删除时会根据lazyfree_lazy_expire这个参数去判断是执行同步删除还是异步删除,另外一点是对于 slave,是不需要执行的,因为会在 master 过期时向 slave 发送 del 指令。
光采用这个策略会有什么问题呢,假如一些key 一直未被访问,那这些 key 就不会过期了,导致一直被占用着内存,所以 redis 采取了懒汉式过期加定期过期策略,定期策略是怎么执行的呢

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
/* This function handles 'background' operations we are required to do
* incrementally in Redis databases, such as active key expiring, resizing,
* rehashing. */
void databasesCron(void) {
/* Expire keys by random sampling. Not required for slaves
* as master will synthesize DELs for us. */
if (server.active_expire_enabled) {
if (server.masterhost == NULL) {
activeExpireCycle(ACTIVE_EXPIRE_CYCLE_SLOW);
} else {
expireSlaveKeys();
}
}

/* Defrag keys gradually. */
activeDefragCycle();

/* Perform hash tables rehashing if needed, but only if there are no
* other processes saving the DB on disk. Otherwise rehashing is bad
* as will cause a lot of copy-on-write of memory pages. */
if (!hasActiveChildProcess()) {
/* We use global counters so if we stop the computation at a given
* DB we'll be able to start from the successive in the next
* cron loop iteration. */
static unsigned int resize_db = 0;
static unsigned int rehash_db = 0;
int dbs_per_call = CRON_DBS_PER_CALL;
int j;

/* Don't test more DBs than we have. */
if (dbs_per_call > server.dbnum) dbs_per_call = server.dbnum;

/* Resize */
for (j = 0; j < dbs_per_call; j++) {
tryResizeHashTables(resize_db % server.dbnum);
resize_db++;
}

/* Rehash */
if (server.activerehashing) {
for (j = 0; j < dbs_per_call; j++) {
int work_done = incrementallyRehash(rehash_db);
if (work_done) {
/* If the function did some work, stop here, we'll do
* more at the next cron loop. */
break;
} else {
/* If this db didn't need rehash, we'll try the next one. */
rehash_db++;
rehash_db %= server.dbnum;
}
}
}
}
}
/* Try to expire a few timed out keys. The algorithm used is adaptive and
* will use few CPU cycles if there are few expiring keys, otherwise
* it will get more aggressive to avoid that too much memory is used by
* keys that can be removed from the keyspace.
*
* Every expire cycle tests multiple databases: the next call will start
* again from the next db, with the exception of exists for time limit: in that
* case we restart again from the last database we were processing. Anyway
* no more than CRON_DBS_PER_CALL databases are tested at every iteration.
*
* The function can perform more or less work, depending on the "type"
* argument. It can execute a "fast cycle" or a "slow cycle". The slow
* cycle is the main way we collect expired cycles: this happens with
* the "server.hz" frequency (usually 10 hertz).
*
* However the slow cycle can exit for timeout, since it used too much time.
* For this reason the function is also invoked to perform a fast cycle
* at every event loop cycle, in the beforeSleep() function. The fast cycle
* will try to perform less work, but will do it much more often.
*
* The following are the details of the two expire cycles and their stop
* conditions:
*
* If type is ACTIVE_EXPIRE_CYCLE_FAST the function will try to run a
* "fast" expire cycle that takes no longer than EXPIRE_FAST_CYCLE_DURATION
* microseconds, and is not repeated again before the same amount of time.
* The cycle will also refuse to run at all if the latest slow cycle did not
* terminate because of a time limit condition.
*
* If type is ACTIVE_EXPIRE_CYCLE_SLOW, that normal expire cycle is
* executed, where the time limit is a percentage of the REDIS_HZ period
* as specified by the ACTIVE_EXPIRE_CYCLE_SLOW_TIME_PERC define. In the
* fast cycle, the check of every database is interrupted once the number
* of already expired keys in the database is estimated to be lower than
* a given percentage, in order to avoid doing too much work to gain too
* little memory.
*
* The configured expire "effort" will modify the baseline parameters in
* order to do more work in both the fast and slow expire cycles.
*/

#define ACTIVE_EXPIRE_CYCLE_KEYS_PER_LOOP 20 /* Keys for each DB loop. */
#define ACTIVE_EXPIRE_CYCLE_FAST_DURATION 1000 /* Microseconds. */
#define ACTIVE_EXPIRE_CYCLE_SLOW_TIME_PERC 25 /* Max % of CPU to use. */
#define ACTIVE_EXPIRE_CYCLE_ACCEPTABLE_STALE 10 /* % of stale keys after which
we do extra efforts. */
void activeExpireCycle(int type) {
/* Adjust the running parameters according to the configured expire
* effort. The default effort is 1, and the maximum configurable effort
* is 10. */
unsigned long
effort = server.active_expire_effort-1, /* Rescale from 0 to 9. */
config_keys_per_loop = ACTIVE_EXPIRE_CYCLE_KEYS_PER_LOOP +
ACTIVE_EXPIRE_CYCLE_KEYS_PER_LOOP/4*effort,
config_cycle_fast_duration = ACTIVE_EXPIRE_CYCLE_FAST_DURATION +
ACTIVE_EXPIRE_CYCLE_FAST_DURATION/4*effort,
config_cycle_slow_time_perc = ACTIVE_EXPIRE_CYCLE_SLOW_TIME_PERC +
2*effort,
config_cycle_acceptable_stale = ACTIVE_EXPIRE_CYCLE_ACCEPTABLE_STALE-
effort;

/* This function has some global state in order to continue the work
* incrementally across calls. */
static unsigned int current_db = 0; /* Last DB tested. */
static int timelimit_exit = 0; /* Time limit hit in previous call? */
static long long last_fast_cycle = 0; /* When last fast cycle ran. */

int j, iteration = 0;
int dbs_per_call = CRON_DBS_PER_CALL;
long long start = ustime(), timelimit, elapsed;

/* When clients are paused the dataset should be static not just from the
* POV of clients not being able to write, but also from the POV of
* expires and evictions of keys not being performed. */
if (clientsArePaused()) return;

if (type == ACTIVE_EXPIRE_CYCLE_FAST) {
/* Don't start a fast cycle if the previous cycle did not exit
* for time limit, unless the percentage of estimated stale keys is
* too high. Also never repeat a fast cycle for the same period
* as the fast cycle total duration itself. */
if (!timelimit_exit &&
server.stat_expired_stale_perc < config_cycle_acceptable_stale)
return;

if (start < last_fast_cycle + (long long)config_cycle_fast_duration*2)
return;

last_fast_cycle = start;
}

/* We usually should test CRON_DBS_PER_CALL per iteration, with
* two exceptions:
*
* 1) Don't test more DBs than we have.
* 2) If last time we hit the time limit, we want to scan all DBs
* in this iteration, as there is work to do in some DB and we don't want
* expired keys to use memory for too much time. */
if (dbs_per_call > server.dbnum || timelimit_exit)
dbs_per_call = server.dbnum;

/* We can use at max 'config_cycle_slow_time_perc' percentage of CPU
* time per iteration. Since this function gets called with a frequency of
* server.hz times per second, the following is the max amount of
* microseconds we can spend in this function. */
timelimit = config_cycle_slow_time_perc*1000000/server.hz/100;
timelimit_exit = 0;
if (timelimit <= 0) timelimit = 1;

if (type == ACTIVE_EXPIRE_CYCLE_FAST)
timelimit = config_cycle_fast_duration; /* in microseconds. */

/* Accumulate some global stats as we expire keys, to have some idea
* about the number of keys that are already logically expired, but still
* existing inside the database. */
long total_sampled = 0;
long total_expired = 0;

for (j = 0; j < dbs_per_call && timelimit_exit == 0; j++) {
/* Expired and checked in a single loop. */
unsigned long expired, sampled;

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;
iteration++;

/* If there is nothing to expire try next DB ASAP. */
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, sampling the key
* space 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;
sampled = 0;
ttl_sum = 0;
ttl_samples = 0;

if (num > config_keys_per_loop)
num = config_keys_per_loop;

/* Here we access the low level representation of the hash table
* for speed concerns: this makes this code coupled with dict.c,
* but it hardly changed in ten years.
*
* Note that certain places of the hash table may be empty,
* so we want also a stop condition about the number of
* buckets that we scanned. However scanning for free buckets
* is very fast: we are in the cache line scanning a sequential
* array of NULL pointers, so we can scan a lot more buckets
* than keys in the same time. */
long max_buckets = num*20;
long checked_buckets = 0;

while (sampled < num && checked_buckets < max_buckets) {
for (int table = 0; table < 2; table++) {
if (table == 1 && !dictIsRehashing(db->expires)) break;

unsigned long idx = db->expires_cursor;
idx &= db->expires->ht[table].sizemask;
dictEntry *de = db->expires->ht[table].table[idx];
long long ttl;

/* Scan the current bucket of the current table. */
checked_buckets++;
while(de) {
/* Get the next entry now since this entry may get
* deleted. */
dictEntry *e = de;
de = de->next;

ttl = dictGetSignedIntegerVal(e)-now;
if (activeExpireCycleTryExpire(db,e,now)) expired++;
if (ttl > 0) {
/* We want the average TTL of keys yet
* not expired. */
ttl_sum += ttl;
ttl_samples++;
}
sampled++;
}
}
db->expires_cursor++;
}
total_expired += expired;
total_sampled += sampled;

/* 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. */
if ((iteration & 0xf) == 0) { /* check once every 16 iterations. */
elapsed = ustime()-start;
if (elapsed > timelimit) {
timelimit_exit = 1;
server.stat_expired_time_cap_reached_count++;
break;
}
}
/* We don't repeat the cycle for the current database if there are
* an acceptable amount of stale keys (logically expired but yet
* not reclained). */
} while ((expired*100/sampled) > config_cycle_acceptable_stale);
}

elapsed = ustime()-start;
server.stat_expire_cycle_time_used += elapsed;
latencyAddSampleIfNeeded("expire-cycle",elapsed/1000);

/* Update our estimate of keys existing but yet to be expired.
* Running average with this sample accounting for 5%. */
double current_perc;
if (total_sampled) {
current_perc = (double)total_expired/total_sampled;
} else
current_perc = 0;
server.stat_expired_stale_perc = (current_perc*0.05)+
(server.stat_expired_stale_perc*0.95);
}

执行定期清除分成两种类型,快和慢,分别由beforeSleepdatabasesCron调用,快版有两个限制,一个是执行时长由ACTIVE_EXPIRE_CYCLE_FAST_DURATION限制,另一个是执行间隔是 2 倍的ACTIVE_EXPIRE_CYCLE_FAST_DURATION,另外这还可以由配置的server.active_expire_effort参数来控制,默认是 1,最大是 10

1
2
onfig_cycle_fast_duration = ACTIVE_EXPIRE_CYCLE_FAST_DURATION +
ACTIVE_EXPIRE_CYCLE_FAST_DURATION/4*effort

然后会从一定数量的 db 中找出一定数量的带过期时间的 key(保存在 expires中),这里的数量是由

1
2
3
4
5
6
7
8
config_keys_per_loop = ACTIVE_EXPIRE_CYCLE_KEYS_PER_LOOP +
ACTIVE_EXPIRE_CYCLE_KEYS_PER_LOOP/4*effort
```
控制,慢速的执行时长是
```C
config_cycle_slow_time_perc = ACTIVE_EXPIRE_CYCLE_SLOW_TIME_PERC +
2*effort
timelimit = config_cycle_slow_time_perc*1000000/server.hz/100;

这里还有一个额外的退出条件,如果当前数据库的抽样结果已经达到我们所允许的过期 key 百分比,则下次不再处理当前 db,继续处理下个 db

在Java8的stream之前,将对象进行排序的时候,可能需要对象实现Comparable接口,或者自己实现一个Comparator,

比如这样子

我的对象是Entity

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
public class Entity {

private Long id;

private Long sortValue;

public Long getId() {
return id;
}

public void setId(Long id) {
this.id = id;
}

public Long getSortValue() {
return sortValue;
}

public void setSortValue(Long sortValue) {
this.sortValue = sortValue;
}
}

Comparator

1
2
3
4
5
6
7
8
9
10
11
12
13
14
public class MyComparator implements Comparator {
@Override
public int compare(Object o1, Object o2) {
Entity e1 = (Entity) o1;
Entity e2 = (Entity) o2;
if (e1.getSortValue() < e2.getSortValue()) {
return -1;
} else if (e1.getSortValue().equals(e2.getSortValue())) {
return 0;
} else {
return 1;
}
}
}

比较代码

1
2
3
4
5
6
7
8
9
10
11
12
13
private static MyComparator myComparator = new MyComparator();

public static void main(String[] args) {
List<Entity> list = new ArrayList<Entity>();
Entity e1 = new Entity();
e1.setId(1L);
e1.setSortValue(1L);
list.add(e1);
Entity e2 = new Entity();
e2.setId(2L);
e2.setSortValue(null);
list.add(e2);
Collections.sort(list, myComparator);

看到这里的e2的排序值是null,在Comparator中如果要正常运行的话,就得判空之类的,这里有两点需要,一个是不想写这个MyComparator,然后也没那么好排除掉list里排序值,那么有什么办法能解决这种问题呢,应该说java的这方面真的是很强大

看一下nullsFirst的实现

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
final static class NullComparator<T> implements Comparator<T>, Serializable {
private static final long serialVersionUID = -7569533591570686392L;
private final boolean nullFirst;
// if null, non-null Ts are considered equal
private final Comparator<T> real;

@SuppressWarnings("unchecked")
NullComparator(boolean nullFirst, Comparator<? super T> real) {
this.nullFirst = nullFirst;
this.real = (Comparator<T>) real;
}

@Override
public int compare(T a, T b) {
if (a == null) {
return (b == null) ? 0 : (nullFirst ? -1 : 1);
} else if (b == null) {
return nullFirst ? 1: -1;
} else {
return (real == null) ? 0 : real.compare(a, b);
}
}

核心代码就是下面这段,其实就是帮我们把前面要做的事情做掉了,是不是挺方便的,小记一下哈