1. HBase：
该模型用于提供对大量结构化数据的随机访问。它建立在 hadoop 文件系统的顶部，本质上是面向列的。它用于将数据存储在 HDFS 中。它是提供数据复制的开源数据库。 HBase 的三个重要组件是 HMaster、Region 服务器、Zookeeper。
2. 卡桑德拉：
Cassandra 旨在处理跨不同商品服务器的大量数据，提供高可用性而不会出现任何故障。它具有能够处理大量数据的分布式架构。数据被放置在具有多个复制因子的不同机器上，以实现高可用性而不会出现任何类型的故障。
HBase 和 Cassandra 的区别：
| HBase | Cassandra |
|---|---|
| HBase is based on Bigtable. | Cassandra is based on DynamoDB. |
| It is written in JAVA. | It is also written in JAVA. |
| It has distributed database. | It has decentralized database. |
| It has triggers. | It doesn’t have triggers. |
| It has no secondary indexes. | It has restricted secondary indexes. |
| Uses a selectable replication factor. | Uses a selectable replication factor. |
| HBase provide more consistency. | It provide less consistency. |
| HBase can handles upto 1000 nodes. | It can handles upto 400 nodes. |
| HBase provide better performance than cassandra. | It has less performance as compared to HBase. |
| HBase is less complex than Cassandra. | It is more complex than HBase. |