Simpler coding than the fixed-size buffer approach
比固定大小的缓冲区方法更简单的代码编写
When this approach is in use, hashed rows destined for a given partition are placed in a buffer and sent to that partition as a group instead of one at a time.
如果使用了这种方法,定向于某个给定分区的散列行首先会进入一个缓冲区中,然后成组地发送到那个分区,而不是一次发送一行。
Unfortunately, this approach itself can become a performance bottleneck if the size of the data requested is considerably larger than the kernel buffer size.
不幸的是,如果所需数据量远大于内核缓冲区大小的话,这个方法本身可能成为一个性能瓶颈。
应用推荐