為了提供Fault Tolerance，Hadoop Namenode在獲取rack id後，會將Block復制幾份存放在不同的rack中。該算法稱為Rack Awreness, 如下圖

HDFS 寫流程：

1. 建立pipline

2. Streaming Data

3.Shutdown and Acknoledgement:

多Blocks的並行操作：

For Block A: 1A -> 2A -> 3A -> 4A
For Block B: 1B -> 2B -> 3B -> 4B -> 5B -> 6B

HDFS Read:

https://www.edureka.co/blog/apache-hadoop-hdfs-architecture/#datanode

https://www.coursera.org/learn/big-data-essentials/lecture/JmzZr/block-and-replica-states-recovery-process-1

Hadoop(2): Blocks存儲管理及讀寫