Persistent, in-memory relational database
- SQL-92 entry-level support, with selected features of SQL-98 and SQL 2003
- Full ACID support
- Multiple table types: persistent, transient, temporary; disk-based and in-memory
- Concurrency control with multiple transaction isolation levels
- Snapshot-consistent checkpointing, transaction logging and automatic roll-forward recovery
- Advanced, cost-based query optimizer
- Multithreaded database engine that leverages multiprocessor and multi-core architectures
- Tables, views, indices and sequences
- Dynamically controlled integrity constraints including primary and foreign keys
- Programmable through SQL and C-language–based stored procedures, triggers and user configurable events
- Configurable security through user and role privileges
- Full Unicode support for all character data, with external encoding separately selectable for each application
- Tools for diagnostics and performance monitoring
- Tools for bulk-loading and exporting data
- Transaction log reader
- Persistent audit trail functionality
- Database compression
- JSON datatype and functions for storing, updating, searching and indexing JSON data
Performance
- Sustainable latencies measured in microseconds for read transactions
- Sustainable latencies measured in tens of microseconds for durable write transactions
- Linear vertical scalability up to number of cores with read transactions
Automatic high availability with solidDB Hotstandby
- Uptimes of up to 99.999% with sub second failover on system failures
- Active/Standby database setup. In case of system failure the Standby database can be transparently switched to Active continuing the database operations without service interruption
- Interfaces to attach your own HA controlling software
- solidDB High Availability Controller, which automatically handles the database failover operations
- ODBC/JDBC transparent connectivity extension, which abstracts away the underlying database server configuration in both normal operational mode and in failure scenarios
Replication with solidDB C-Replication
- High performance reliable asynchronous replication
- Replicate within datacenter or geographically
- One to many, many to one, bi-directional and hierarchical replication topologies
- Can be combined with solidDB Hotstandby for high availability
Embeddability
- Small footprint and hybrid in-memory disk solution
- Ability to be deployed as a server process to enable access by applications over the network
- Provided also as a dynamically loadable library to be linked with Java or C/C++ applications that are deployed on the same server as solidDB to boost performance and to make solidDB invisible for embedded deployments
- Shared memory access for co-located applications
- Embedding solidDB does not restrict other capability deployments like HSB and CREP
solidDB Grid (new in solidDB 201.0)
- Highly available database solution that scales in terms of database size and transaction performance
- Appears to the user as a single logical database server
- Can be up- and down scaled dynamically at runtime
- Transactional performance scales linearly with its size
- Continuous service and data integrity even in the event of node or network failures
- Grid driver provides a logical connection to the Grid, eliminating the need for the user to establish connections to individual servers operating within the Grid
- Global synchronized database schema
- Common administrative tools
- Automated node membership management
Vector datatype and functions (new in solidDB 201.0)
- Efficient storage, indexing, and retrieval of vector data
- Store floating-point arrays and leverage automatic indexing
- Distance functions like Euclidean distance and dot product for flexible similarity measurements
