[high concurrency] interviewer

Use Java Heap memory to store objects . The advantage of using heap caching is that there is no serialization / Deserialization , Is the fastest cache . The disadvantages are obvious , When the amount of cached data is large ,GC（ Garbage collection ） The pause time will be longer , Storage capacity is limited by the size of heap space . Generally through soft reference / Weak references are used to store cached objects . That is, when the heap memory is low , This part of memory can be forcibly reclaimed to free up heap memory space . Generally, heap cache is used to store hot data . have access to Guava Cache、Ehcache 3.x、 MapDB Realization .

[](() Out of heap memory

That is, the cached data is stored in off heap memory , Can reduce the GC Pause time （ Move heap objects out of the heap ,GC Fewer objects are scanned and moved ）, Can support more cache space （ Only limited by machine memory size , Not affected by heap space ）. however , Serialization is required when reading data / Deserialization . therefore , It will be much slower than heap cache . have access to Ehcache 3.x、 MapDB Realization .

[](() Disk caching

That is, cache data is stored on disk , stay JVM The data still exists when restarting , And heaps / Out of heap cache data will be lost , Need to reload . have access to Ehcache 3.x、MapDB Realization .

[](() Distributed cache

Distributed caching can use ehcache-clustered( coordination Terracotta server) Realization Java Inter process distributed cache . You can also use Memcached、Redis Realization .

When using distributed caching , There are two modes as follows ：

• standalone mode ： Store the hottest data to the heap cache , Relatively hot data to the off heap cache , Not hot data to disk cache .

• Cluster pattern ： Store the hottest data to the heap cache , Relatively hot data to external cache , Full data to distributed cache .

[](() Cache recovery policy

------------------------------ 《 A big factory Java Analysis of interview questions + Back end development learning notes + The latest architecture explanation video + Practical project source code handout 》 Free open source Prestige search official account 【 Advanced programming 】 ----------------------------------

The cache recycling strategy generally includes ： Space based recycling strategy 、 Based on capacity （ Space ） Recycling strategy for 、 Time based recycling strategy and object reference based recycling strategy .

[](() Based on space

Storage space is set based on the cache , If set to 10MB, When the maximum storage space is reached , Remove data according to certain policies .

[](() Based on capacity

The maximum size of the cache is set based on the capacity , When the cached entry exceeds the maximum size , Remove old data according to certain policies .

[](() Based on time

TTL(Time To Live): Survival period , That is, a period of time from creation to expiration of cached data （ Whether or not visited during this time period , All cached data will expire ）.

TTI(Time To Idle)： Idle period , That is, the time to remove the cache after the cached data has not been accessed .

[](() Based on object reference

Soft citation ： If an object is a soft reference , Then when JVM When the heap is low on memory , The garbage collector can recycle these objects . Soft references are suitable for caching , So as to be JVM When the heap is low on memory , You can recycle these objects to make room for strongly referenced objects , To avoid OOM.

Weak reference ： When the garbage collector reclaims memory , If weak references are found , Then recycle it immediately . Relative to soft reference , Weak references have a shorter life cycle .

Be careful ： Only when no other strong reference object references weak references / When soft referencing objects , The reference is recycled only when garbage collection . That is, if there is an object （ It's not a weak quote / Soft reference objects ） A weak quotation was quoted / Soft reference objects , Then the weak reference will not be recycled during garbage collection / Soft reference objects .