Raima Database Manager can be divided up so that part of the database resides in main memory while the rest of the database remains on-disk. This is referred to as the hybrid database model. Typically data that is constantly changing and repeatedly accessed is placed in main memory while more consistent and/or aggregated data is stored on disk. In other cases data that is easily recreated like indexes are kept in main memory providing fast lookup performance while the actual data is written to disk.

A key concern is often durability of data when implementing an in-memory database. As main memory is more volatile than persistent disk storage, certain mechanisms need to be in place to ensure ACID compliance. A significant amount of engineering has gone into the Raima database products to provide high speed access to traditional, on-disk databases while ensuring transaction consistency and data integrity.

RDM’s in-memory function uses transaction logs that can be written to disk providing additional rollback and recovery, and fewer lost transactions in the case of a system failure. The in-memory functionality can be used with full ACID transactional integrity support or, where much higher performance is needed and lower DBMS-managed integrity requirements are acceptable, ACID enforcement can be relaxed.

Main memory is more expensive than persistent disk storage, hence an in-memory database needs to have a compact design to utilize available storage more efficiently. RDM is small footprint, cross-platform database designed for distributed architecture in resource-constrained environments that can be configured to run completely in memory.

Traditionally, a database that relies on main memory will provide much faster access to existing data, but may also be more vulnerable to loss if there is a program or computer error. To safeguard against this, RDM asks the operating system to sync a file at key moments, ensuring that the data is safe no matter when a computer may fail.

The sync operation (synchronize to disk) will not return control to the program until the file contents exist on the permanent media. Raima Database Manager is designed around the use of the Transactional File Server (TFS) which manages all database input and output along with the multi-version concurrency control (MVCC) implementation.

In-memory support provides the highest possible database performance. Applications that require instant response times and maximum efficiency will benefit greatly from implementing an in memory database. The same applies to diskless systems and devices. Raima tests shows performance results up to 300% by comparing the times required to do the same database operations on a disk-based database with an in-memory version of the same database.

RDM is today used worldwide in a wide range of data management solutions within in industrial automation systems, military flight control systems, telecom routers & switches, financial trading systems, medical equipment, data backup solutions, consumer electronic devices and more.