# 
HBase key capabilities and features combing

By [rojeck](https://paragraph.com/@rojeckluo) · 2021-12-19

---

    HBase is playing an increasingly important role in real-time query due to its high performance in storage and reading and writing, but because it belongs to the NOSQL database type, it is not suitable for relational data. HBase query can only be queried by its rowkey (we can think of it as the unique index of the table in HBase). Therefore, the design of rowkey is particularly important in the design of using HBase. In addition, the design of rowkey is also related to the storage location of data in the database. If the design of rowkey is improper, it will cause data hotspot problems in HBase partitions, that is, centralized data access. In a certain node or region, it will eventually lead to performance degradation or the region is unavailable due to heavy load.
    
    In order to prevent data hotspots during write operations, data should be written to multiple regions as much as possible at the same time when designing rowkeys. Here are several common rowkey design methods.

---

*Originally published on [rojeck](https://paragraph.com/@rojeckluo/hbase-key-capabilities-and-features-combing)*