Apache Ignite 3.0

This page describes ideas, thoughts, and planning around Apache Ignite 3.0 release.

Description

See release page: Apache Ignite 3.0.0

Development Process

All the development for the Apache Ignite 3.x happens in the dedicated repository: https://github.com/apache/ignite-3.

Major Changes and Features

Improved Architecture

Ignite 3 introduces a new transactional protocol and strictly serializable model that provides superior transactional consistency and simplifies distributed transactional application development.

It also makes use of the RAFT consensus algorithm to manage cluster nodes and partition consistency. With RAFT, several nodes form a cluster management group, automatically managing and monitoring cluster status. This approach provides improved stability for the cluster, as well as out-of the box protection from split-brain.

Additionally, the pluggable storage feature gives developers the flexibility to choose the storage engine based on application needs.

Management Tools

Ignite 3 introduces management tool that can communicate with any node in the cluster to seamlessly apply the configuration, and provides you with an option to work in interactive mode with command autocomplete. This new tool is also provided as a separate distribution, so that it can be set up on any machine that has access to the node address.

Improved Usability

Ignite 3 provides bundled installation packages, allowing you to easily deploy on any container platform or operating system, on-premises or in the cloud.

Cluster and node configuration is reworked in HOCON based format. This allows for much more light-weight configuration files that are also easier to edit. To make things even easier, configuration is now split between cluster and node configuration.

Enhanced SQL Support

Ignite 3 prioritizes SQL for interacting with data storage. Transactions are now supported seamlessly across all APIs, including SQL. Strictly serializable SQL transaction capability enables application migration from traditional RDBMS platforms.

Data schema management is also made more uniform across different APIs in Ignite 3. Now SQL, Key-Value API, and the new Record API all share access to the same schema, making it easy to access the same data from applications with different data models.