GridGain can function as a distributed in-memory database that supports data processing APIs including SQL, key-value, compute, machine learning and more. The GridGain in-memory data grid distributes your dataset across a cluster of servers while the distributed SQL capabilities allows you to read and write to the database using standard database commands. The system provides ACID transaction guarantees and supports ANSI-99 SQL including DDL and DML.
The GridGain in-memory database delivers 1,000 times faster performance than disk-based databases because your data is stored and processed in RAM. GridGain is deployed on a distributed cluster of servers so you can easily scale out the GridGain in-memory database by adding additional nodes to the cluster - the system rebalances your data automatically.
In-Memory Database Built on a Memory-Centric Architecture
The memory-centric GridGain architecture allows you to execute distributed SQL, key-value and other operations across different memory layers. If your organization deploys a variety of memory technologies such as DRAM, non-volatile memory, and 3D XPoint, you can tune the configuration of your system to use a combination of memory options which provides the best trade off between price and performance for your organization.
You can use the optional Persistent Store feature to achieve the performance and scale of in-memory computing, the durability of disk, and strong consistency, all in one system. The Persistent Store feature maintains a copy of your in-memory database on disk which serves as a backup for data recovery purposes. GridGain can also transact and query data whether it is in-memory or on disk so you can use your in-memory database immediately upon restart by transacting against the data on disk while your full dataset is being loaded into memory.