Apache Ignite can be used as an in-memory data grid that accelerates and scales your existing databases and data stores. The grid supports key-value and ANSI SQL APIs, ACID transactions, co-located processing, and machine learning libraries. As an in-memory data grid, Ignite is frequently used to increase the performance and scalability of real-time applications, to support high-performance computing, and to cache data that is scattered across databases.
What is an In-Memory Data Grid?
An in-memory data grid is an advanced read-through/write-through cache that is deployed on top of multiple databases. The grid supports various APIs, such as SQL, compute, and key-value. Applications write to and read from the grid, and the grid propagates changes to the underlying data stores in a consistent way.
With Ignite, your applications can utilize multiple APIs for data processing. Key-value and SQL queries enable you to request, join, and group distributed datasets. Compute and machine-learning APIs execute logic close to the data, thus eliminating expensive data shuffling over the network.
IGNITE AS AN IN-MEMORY DATA GRID AND EXTERNAL DATABASES
Ignite can improve the performance and scalability of most external databases and data stores (such as RDBMS, NoSQL, and Hadoop) by sliding in as an in-memory data grid between the application and the database layers. When the application writes data to the cache, Ignite writes-through or writes-behind all data modifications to the underlying external store. Ignite also performs ACID transactions (coordinating and committing transactions across both the cluster and the database).
Additionally, Ignite can be deployed as a shared and unified in-memory layer that stores datasets that originate from disjointed databases. Your applications can consume all the data from Ignite as a single store while Ignite synchronizes the original databases whenever in-memory data is updated.
IGNITE NATIVE PERSISTENCE
Ignite Native Persistence is a distributed ACID-compliant, SQL-compliant disk store that transparently integrates with the Ignite in-memory layer. When Native Persistence is enabled, Ignite stores both data and indexes on disk, thus eliminating the time-consuming cache warm-up step. Native Persistence keeps a full copy of data on disk, so you are free to cache a subset of records in memory. If a required data record is missing from memory, Ignite reads the record from the disk automatically, regardless of the API (whether SQL, key-value, or scan queries).