Massive Database Performance Gains with In-Memory Computing
The GridGain® in-memory computing platform enables users to build modern, scalable, high performance applications. GridGain is inserted as a distributed computing layer between the application and data layers. Application see greatly improved database performance of up to 1,000x because the full data set is held in RAM, eliminating the need to read data from disk before processing. The system can easily be scaled out by adding more nodes to the cluster to increase the available RAM and CPU pool and allow you to maintain petabytes of data in-memory. Deployed as an in-memory database using the GridGain Persistent Store feature, you can use GridGain as your database of record for new applications while having processing access to the full dataset on disk and holding a user-definable amount of data in-memory.
The GridGain in-memory computing solution can improve the performance of your existing SQL, NoSQL and Hadoop databases by up to 1,000x without replacing your existing data stores. GridGain can both accelerate your database and provide you with massive database scalability for your application.
GridGain includes an in-memory data grid, in-memory database, streaming analytics, and a continuous learning framework for real-time machine and deep learning. GridGain features ANSI-99 SQL support, including DDL and DML, and ACID transaction guarantees. Native integrations with many open source solutions and databases are also part of the platform.
GridGain for Database Performance
- Distributed PostgreSQL - GridGain automatically distributes your database across the nodes of the GridGain cluster, providing scalability, high performance, and high availability which is not possible using standard Postgres
- Distributed MySQL - GridGain automatically distributes your MySQL database over the nodes of the GridGain cluster to provide high availability, high performance, and massive scalability
- Redis Alternative - GridGain is a powerful alternative to Redis because it includes many additional features not found in Redis that are highly valuable when moving to in-memory computing
- Cassandra Alternative - GridGain offers the same scalability and high availability capabilities as Cassandra but also offers ACID transaction guarantees, ANSI-99 SQL support, and in-memory computing for lightning fast performance
- Cassandra Transactions - Inserting GridGain between Apache Cassandra and an application provides ACID transactions for Cassandra plus improves performance
- Ad Hoc Cassandra SQL Queries - the portion of your Cassandra data which is loaded in GridGain is indexed and can be queried using ANSI-99 SQL
- IoT Database Acceleration - GridGain is deployed for IoT database acceleration between the application and data layers and supports ACID transactions and distributed ANSI-99 SQL including DDL and DML
GridGain for Big Data Applications
- Persistent Spark RDDs - Spark RDDs are stored in the in-memory GridGain storage, available to share state between Spark jobs and processes
- Apache Spark SQL Acceleration - GridGain can store and index the Spark data set in memory so SQL queries can run much faster compared to running a full scan when executing SQL against the un-indexed Spark data
- Hadoop Acceleration - the Ignite file system (IGFS) can be easily substituted for HDFS which moves your data to memory where it can be processed up to 3x faster than from disk