There was a time when we talked about Apache Hive limits, and the list went on and on, but that’s no longer the case. Apache Hive is gradually overcoming its limits and transforming its appearance.
Hortonworks rallied around Hive and began adding features such as transactional tables, ACID support, index support, and LLAP support, among others.
- Previously, row level inserts and changes were not supported. However, as of version 0.14, the answer has been changed from NO to Yes, despite the fact that it necessitates additional table properties and configurations.
- Atomity, consistency, and durability were given at the partition level until Hive 0.13. Turning on one of the available locking mechanisms could provide isolation (ZooKeeper or in memory). With the addition of transactions, full ACID semantics at the row level are now possible, permitting one application to add rows while another reads from the same partition without interfering.
This means, Hive now supports full ACID attributes, allowing us to use transactions, create transactional tables, and perform queries like Insert, Update, and Delete on tables starting with version 0.14. - Until Hive 0.7, indexing was not supported. However, index support is included in subsequent versions, up to 3.0. Since 3.0, however, indexing has been eliminated, and alternative solutions like as materialized views and columnar file formats (Parquet, ORC) have been added, which do selective scanning and may even skip entire files/blocks from being read.
- Low Latency Analytical Processing (LLAP), a new component introduced in Hive 2.0 that makes Hive faster by employing persistent query infrastructure and enhanced data caching, allowing interactive and subsecond SQL. Hive SQL queries and data types are fully compatible with LLAP.
Hope you find this article helpful.
Please do subscribe for more interesting updates.
Big Data and SQL 