Cassandra Gets an Indexing Upgrade
A new data storage framework for indexing on Apache Cassandra released this week by DataStax is aimed at easing adoption of cloud-native NoSQL databases by eliminating developers’ tradeoffs between scale, stability and performance.
Cassandra users increasingly employ indexes as a means other than a partition key to access data on Cassandra. (A table partition key is an ordered set of columns in a table.) Secondary indexes are also used to query tables that are normally not accessible. The new DataStax approach, dubbed storage-attached indexes, is designed to simplify the filtering of data via a new secondary index.
The result, DataStax said, is a simpler way for application developers to access data on Cassandra NoSQL databases. Storage-attached indexing “eliminates many of these tradeoffs, making development and data modeling in Apache Cassandra easier to use, while also increasing stability and performance and giving architects and operators fewer moving parts to manage,” said Ed Anuff, chief product officer at DataStax.
The Cassandra indexing implementation available on the DataStax Astra and enterprise platforms allows developers to access multiple columns on the same table. Expanded storage resources are designed to boost scaling and stability on NoSQL databases while reducing storage disk requirements.
For example, promoters note that storage-attached indexing shares data common among multiple column indexes on the same table.
DataStax, Santa Clara, Calif., said the new indexing approach is available now on the Astra platform as well as the 6.8.3 version of its enterprise distribution of Cassandra.
The company also said this week is has launched a new “Cassandra Enhancement Proposal” for sharing the new storage-attached framework as a new secondary index for Cassandra distributed databases. Among the goals, promoters said, is avoiding scaling problems while providing a means of indexing multiple columns on the same table.
The storage-attached indexing effort represents another step by DataStax in efforts to attract application developers to its flagship Astra platform, which migrated to the public cloud earlier this year. Combining Cassandra’s inherent scalability with ease-of-use is seen as a way of winning over developers when selecting databases for new distributed applications.
The new indexing tool would, for example, allow for more efficient data queries for a given condition.