We believe in continuously investing in our SimbaEngine SDK software, to provide benefit for our valued customers. In our latest release, SimbaEngine X SDK, v 10.1.1, we have focused on improvements to our Simba SQL Engine.
The Simba SQL Engine is a high-performance, standard-compliant SQL parser and execution engine. We designed it to allow our customers to build drivers for data sources that are not fully SQL compliant. The SQL engine is available in the C++ and Java.
Stored Procedure Support in the Java SQL Engine
A stored procedure is an operation that allows the DSI to take advantage of internal functions or extended non-SQL functionality that the data store may support. Custom functions that can be implemented inside these stored procedures may allow access to data that is not stored in standard relational tables. Instead of calling a SQL statement on the data store, an application can call a stored procedure. We have added support for stored procedures to our Java SQL engine. This feature is now available in both our Java and C++ SDK.
Hybrid Hash Enhancement
We recently made a significant improvement in our support for equi-joins on large datasets by using a Simba-optimized version of the hybrid hash algorithm. We have extended this feature to work with collations. Now any specific language collation rules can be implemented by a DSII, and used during join operations. We have also added support for user defined types, so that now, even user defined types, can benefit from the fast performing hasher joins.
Memory Manager Enhancement
We have enhanced the memory manager in the SimbaEngine SDK to provide better concurrency support. When many concurrent queries are sent, the memory manager will ensure that they wait until enough memory becomes available, thereby ensuring that they will not fail due to “out of memory error”. The calling application now also has full control on when to cancel slow queries. This offers more flexibility for the calling application.
Download SimbaEngine X SDK 10.1.1 to try it out!