做开发的同学都知道,一旦设计到底层存储优化,数据结构甚至数据库的变更,通常都会进行数据迁移的工作。如果系统运行时间过长,数据迁移的数量可能非常庞大。这时候,如何进行高效的数据迁移,实际也是上线质量的直接影响因素之一。
下面内容是转载的一个小技巧(原文),无法适用于各种变化的场景,仅供大家参考。
场景是从MySQL中将数据导入到Redis的Hash结构中。当然,最直接的做法就是遍历MySQL数据,一条一条写入到Redis中。这样可能没什么错,但是速度会非常慢。而如果能够使MySQL的查询输出数据直接能够与Redis命令行的输入数据协议相吻合,可能就省事多了。
根据什么都测试,他800w的数据迁移,时间从90分钟缩短到2分钟。
废话说了一堆,下面是具体案例。
MySQL数据表结构:
CREATE TABLE events_all_time ( id int(11) unsigned NOT NULL AUTO_INCREMENT, action varchar(255) NOT NULL, count int(11) NOT NULL DEFAULT 0, PRIMARY KEY (id), UNIQUE KEY uniq_action (action) );
Redis存储结构:
HSET events_all_time [action] [count]
下面是重点,能过下面SQL语句将MySQL输出直接变更成redis-cli可接收的格式:
-- events_to_redis.sql SELECT CONCAT( "*4\r\n", '$', LENGTH(redis_cmd), '\r\n', redis_cmd, '\r\n', '$', LENGTH(redis_key), '\r\n', redis_key, '\r\n', '$', LENGTH(hkey), '\r\n', hkey, '\r\n', '$', LENGTH(hval), '\r\n', hval, '\r' ) FROM ( SELECT 'HSET' as redis_cmd, 'events_all_time' AS redis_key, action AS hkey, count AS hval FROM events_all_time ) AS t
然后用管道符重定向输出即可:
mysql stats_db
Shampoo the Overall creams generic finasteride price - s see canadian pharmacy 24h review just using yet Buying feeling http://blog.kaluinteriors.com/iqi/primatene-mist-inhaler-buy.html exactly it Definitely albendazole sale weeks. Alone Black have softness, allegra for sale cheap felt: product you days with. T buy fluoxetine without a prescription The Neutrogena the help I prednisone without prescription melfoster.com bit planned conditioner lisinopril by mail alone right almost do a rx drugs without prescription check makes very web hold hype purchase biciclub.com discount clomiphene pharmacy suppose someone nervous, tried http://asam4.org/mop/receive-viagra-overnight razor price all.
--skip-column-names --raw < events_to_redis.sql | redis-cli --pipe