更糟的是，在堆上分配对象时，缓存遗漏还有一个特别讨厌的内存交互。
Worse, a cache miss when allocating an object on the heap has a particularly nasty memory interaction.
新对象是从这个高速缓存的末端分配的，而不必获取堆锁，因此效率非常高。
A new object is allocated from the end of this cache without the need to grab the heap lock, so it is very efficient.
所以，在堆上分配对象，比起在堆栈上分配对象，会带来更多缓存遗漏。
As a result, allocating an object on the heap will likely entail more cache misses than allocating that object on the stack.
向Portlet创建的Work对象分配一个密钥，WebSphere使用该密钥在缓存中异步存储它的结果。
The Work object created by the portlet was assigned a key and WebSphere used this key to store its results to the cache asynchronously.
从inode_cacheslab分配器分配单个inode缓存对象。
Individual inode cache objects are allocated from the inode_cache slab allocator.
Jeff的分配器是围绕对象缓存进行的。
使用散列表创建dentry缓存，并且根据对象名分配缓存。
The dentry cache is built using a hash table and is hashed by the name of the object.
如何分配缓存在面向对象的方式方法？
如何分配缓存在面向对象的方式方法？
应用推荐