Redis使用LUA脚本计算HashMap中的Value的SUM

统计Redis中HashMap缓存中Value的数据值:

 

local  sizeList = redis.call("HVALS", KEYS[1])
local  total_size = 0
for    index, line in pairs(sizeList) 
do 
       total_size = total_size + tonumber(line)
end
return total_size

 

1. 上传脚本

[eagle@greatwqs src]$ ./redis-cli script load "$(cat /home1/eagle/apps/redis/lua/folder_size.lua)"

"104fc275456ac249b6e266a345d424cf9efe62f4"

2. 造数据

[eagle@greatwqs src]$ ./redis-cli

127.0.0.1:6379> HMSET wqstest folder-1 100 folder-2 200 folder-3 300

OK

127.0.0.1:6379> hvals wqstest

1) "100"

2) "200"

3) "300"

127.0.0.1:6379> exit

3. 使用脚本测试数据

[eagle@greatwqs src]$ ./redis-cli --eval /home1/eagle/apps/redis/lua/folder_size.lua wqstest

(integer) 600

 

PS: 使用脚本造数据

local  redisKey = KEYS[1]
redis.call('DEL', redisKey) 
local total_count = tonumber(KEYS[2])

for loopIndex=1, total_count do
  redis.call('HMSET', redisKey, loopIndex, loopIndex) 
end
return redis.call('HLEN', redisKey) 

[eagle@greatwqs src]$ ./redis-cli --eval /home1/eagle/apps/redis/lua/create_hm_value.lua test 500000

(integer) 500000

造50W条数据, 耗时感觉在秒级别.

 

再用上面第一个脚本计算一下SUM:

[eagle@greatwqs src]$ ./redis-cli --eval /home1/eagle/apps/redis/lua/folder_size.lua test

(integer) 125000250000

很快, 低于500MS就返回了.