Redis memory optimization and distributed locking

One ： Memory optimization .

（1） stay key In a large number of cases , Try to replace the string with a small hash table .

（2） because redis Default in hash key No more than 512,value No more than 64 Byte time , Use ziplist.

（3） Additional development is needed sharding function , To break up the big hash table .

The problem is coming. , In string form , Every key You can set the expiration time separately . Hash table does not support child key Set expiration , To do ？

terms of settlement ： stay value Save time stamp in . Offline script scheduled scan .

This method is used in the case of a large amount of data , Memory can be compressed 3 About times .

Two ： Distributed lock （ single redis example ）

Distributed lock acquisition ：set key value ex(30) nx

Release of distributed lock ：（1） Use eval perform lua Scripts compare values . Delete when equal . Use lua Scripts are meant to ensure atomicity .

                             （2） If the task execution exceeds 30 What to do after seconds ？ You can open threads to use expire Give him more time to live . Until the current task is completed , To release .