Raijin has announced the release of version 2.0 of its powerful, schemaless SQL-like database engine. This version introduces several performance improvements.
Read on for the highlights and check out the Raijin release notes for a complete list of the features and improvements.
Enhanced table partitioning
Table partitioning is a key factor in database management, improving query performance by only searching through relevant information and optimizing storage by efficiently pruning irrelevant content.
Raijin 2.0 brings a major overhaul to table partitioning, making it more flexible and powerful. By moving from expression-based to multidimensional partitioning, Raijin 2.0 now can partition by time and a table field simultaneously. Working with these two dimensions allows for selective querying and pruning.
Updated OpenAPI specification
The OpenAPI specification now includes the latest Query and Ingestion endpoints. This further enables Raijin’s integration, by providing a more accurate and complete specification information.
Sorting processing optimization
By default, Raijin uses the optimized version of the sorting algorithm to speed up the sorting process. Depending on the sorting query, optimized sorting could potentially require considerable amounts of memory. In those situations, Raijin would be forced stop executing the query if not enough memory was not available.
Now, Raijin will use the optimized sorting up to the memory limit, and then fall back to the standard sorting method. This allows the running of large sort queries without the risk of failing due to memory constraints.
Improved reliability during crash recovery
This version of Raijin also includes improvements to the crash recovery process. There was an issue where, in low disk space situations, heap files would be missing when flushing tables after a crash. Raijin will now recover from crashes without data loss, with heap files available after a crash recovery.
Collect logs with NXLog
As a database engine, Raijin is geared towards semi-structured data, making it a perfect solution for log storage. Log data is notoriously unstructured, making it hard to store in rigid table structures like traditional SQL systems.
Raijin complements NXLog’s powerful log collection and centralization solutions. With NXLog Enterprise Edition’s Raijin integration, you can seamlessly and easily send logs from NXLog Enterprise Edition to a Raijin database instance. The two fit hand in glove, providing the methods of capturing, processing, and storing log data for a complete log management solution.
