本文分享自天翼云开发者社区《Cache写机制Write-through与Write-back》，作者：LeonHao

cache写机制中的write through 和 write back https://en.wikipedia.org/wiki/Cache#Writing_Policies

Write-through- Write is done synchronously both to the cache and to the backing store.
Write-back (or Write-behind) - Writing is done only to the cache. A modified cache block is written back to the store, just before it is replaced.

• Write-through（直写模式）在数据更新时，同时写入缓存Cache和后端存储。此模式的优点是操作简单；缺点是因为数据修改需要同时写入存储，数据写入速度较慢。
• Write-back（回写模式）在数据更新时只写入缓存Cache。只在数据被替换出缓存时，被修改的缓存数据才会被写到后端存储。此模式的优点是数据写入速度快，因为不需要写存储；缺点是一旦更新后的数据未被写入存储时出现系统掉电的情况，数据将无法找回。
• Writearound ats现有cache写入模式

Write-misses写缺失的处理方式

对于写操作，存在写入缓存缺失数据的情况，这时有两种处理方式： Write allocate (aka Fetch on write) - Datum at the missed-write location is loaded to cache, followed by a write-hit operation. In this approach, write misses are similar to read-misses. No-write allocate (aka Write-no-allocate, Write around) - Datum at the missed-write location is not loaded to cache, and is written directly to the backing store. In this approach, actually only system reads are being cached.

Write allocate方式将写入位置读入缓存，然后采用write-hit（缓存命中写入）操作。写缺失操作与读缺失操作类似。

No-write allocate方式并不将写入位置读入缓存，而是直接将数据写入存储。这种方式下，只有读操作会被缓存。

无论是Write-through还是Write-back都可以使用写缺失的两种方式之一。只是通常Write-back采用Write allocate方式，而Write-through采用No-write allocate方式；因为多次写入同一缓存时，Write allocate配合Write-back可以提升性能；而对于Write-through则没有帮助。

处理流程图

Write-through模式处理流程： A Write-Through cache with No-Write Allocation Write-back模式处理流程：