Hyper-Converged Database (HCD) release notes
Hyper-Converged Database (HCD) is compatible with Apache Cassandra® 4.0 and adds additional production-certified changes. HCD includes the following innovations:
-
Storage Attached Indexes (SAI) for advanced database indexes
-
Data API for a simple JSON document-oriented interface to data. For more details on the Data API and how it compares to the CQL interface, see API reference overview.
-
Trie based data structures for significantly improved performance
Here are the features, enhancements, and updates per HCD version.
Version 1.1
This version was released on November 29, 2024 and includes the following items:
New features and enhancements
-
Added the option to upgrade from DataStax Enterprise (DSE) 5.1 to HCD 1.1. This update allows upgrades only from DataStax Enterprise (DSE) 5.1 to HCD 1.1, and not to HCD 1.0.
-
Added an option to bypass the name validation policy, allowing datacenter names for datacenters that do not yet exist. (HCD-18)
Known issues
When upgrading clusters, there are known issues with:
-
User-defined functions (UDFs) and user-defined aggregates (UDAs). (DSP-24606)
-
Schemas containing tuples with freezable inner types. (DSP-24600)
Version 1.0
This version was released on July 10, 2024 and includes the following items:
HCD offers a cloud-native solution for modern data infrastructure. Paired with Mission Control as the control plane, HCD leverages DataStax’s extensive experience managing thousands of databases, delivering the same stability and scale in self-managed environments. The initial release supports Cassandra workloads with notable enhancements, such as the Data API, which eliminates data modeling with Cassandra and has been adopted by developers in Astra DB. For generative AI workloads, DataStax has an add-on for the state-of-the-art vector search engine, JVector.
