Hoard内存分配器的学习手记

准备工作

发现我膜拜的项目中竟然直接应用了hoard的代码555，所以拿来看看到底哪里厉害！

Hoard项目的官方网站在http://www.hoard.org/

它自己配了一部分英文简介，很容易看懂。

The Hoard memory allocator is a fast, scalable, and memory-efficient memory allocator for Linux, Solaris, Mac OS X, and Windows. Hoard is a drop-in replacement formalloc that candramatically improve application performance, especially for multithreaded programs running on multiprocessors and multicore CPUs. No source code changes necessary: just link it in or set one environment variable

另外，它提供了tar包的下载，不过不要选官网上的，到git-hub上clone一个最好，官网上的tar包有redefination错误。

 

简介

这个项目是出自于ASPLOS 2000年的一篇文章 Hoard: A Scalable Memory Allocator for Multithreaded Applications

为了更好理解，我大概看了一下文章。

Hoard: A Scalable Memory Allocator for Multithreaded Applications

摘要：

hoard就是一个为并行多线程C或者是C++程序设计的内存分配器。

之前为串行程序设计的串行分配器的扩展性不行，现有的并行分配器的性能不好，而且有的分配器对堆的管理会引入假共享（false sharing ）

Worse, many allocators exhibit a dramatic increase in memory consumption when confronted with a producer-consumer pattern of object allocation and freeing. This increase in memory consumption can range from a factor of P (the number of processors) to unbounded memory consumption.

这篇文章，自称 Hoard, a fast, highly scalable allocator that largely avoids false sharing and is memory efficient.

而且Hoard combines one global heap and per-processor heaps with a novel discipline that provably bounds memory consumption and has very low synchronization costs in the common case.