Few days ago The Apache Software Foundation announced the release of the new version of Apache Cassandra 4.0 which is a distributed database management system It belongs to the class of noSQL systems and is designed to create highly scalable and reliable storages of large amounts of data stored in the form of an associative array.
This new version of Apache Cassandra 4.0 is considered a stable version so it can be used for production deployments and has already been tested in Amazon, Apple, DataStax, Instaclustr, iland and Netflix infrastructures with clusters of more than 1000 nodes.
Main new features of Apache Cassandra 4.0
This new version of Apache Cassandra 4.0 represents nearly 1,000 bug fixes, improvements, and new features of which the following stand out:
- Increased speed and scalability: transmits data up to 5x faster during scale operations and up to 25% faster performance on reads and writes, providing a more elastic architecture, particularly in cloud and Kubernetes deployments.
- Improved consistency: keeps data replicas in sync to optimize incremental repair for faster and more efficient operation and consistency between data replicas.
- Improved security and observability: the audit trail tracks user access and activity with minimal impact on workload performance. The new capture and playback enables analysis of production workloads to help ensure security and regulatory compliance with SOX, PCI, GDPR, or other requirements.
- New configuration settings: exposed system metrics and configuration settings provide flexibility for operators to ensure they have easy access to data that optimize deployments.
- Minimized latency: garbage collector pause times are reduced to a few milliseconds with no latency degradation as heap size increases.
- Better compression: Enhanced compression efficiency relieves unnecessary stress on disk space and improves read performance.
In addition to this, it is also noted that audit log support to track authentication operations of users and all the CQL queries executed, as well as the ability to maintain a complete binary record of requests, allowing you to save all request and response traffic.
Likewise, too the experimental option to compare all Merkle trees (Merkle Tree) is highlighted. For example, enabling the option on a cluster with 3 nodes, where two replicas are the same and one is outdated, will result in updating the outdated replica using only a copy operation of the current replica.
As well, added support for virtual tables that do not reflect data stored in SSTables, but information displayed through the API (performance metrics, configuration information, cache content, information about connected clients, etc.).
The efficiency of compressed storage has been improved to reduce disk space consumption and improve read performance.
On the other hand, it stands out that Added experimental support for transient replication and Cheap Quorums. Temporary replicas do not store all the data and use incremental recovery to be consistent with full replicas. Lightweight Quorums are write optimizations that do not write to temporary replicas until a sufficient set of full replicas is available.
As for the data related to the space of the system key (system. *), This is now in the first directory by default instead of being distributed among all the data directories, allowing the node to remain operational in the event of a failure of one of the additional disks.
De the other changes that stand out:
- Added experimental support for Java 11.
- Added support for arithmetic operations in CQL queries.
- The "nodetool cfstats" command added support for sorting by certain metrics and limiting the number of lines displayed.
- The settings are provided to restrict user connections to specific data centers only.
- Added the ability to limit the intensity (frequency cap) of operations to create and delete snapshots.
- Python 3 support is implemented in cqlsh and cqlshlib (Python 2.7 support is still preserved).
Finally, if you are interested in knowing more about it of this new version, you can check the details In the following link.