Solving low latency data management problems at very high concurrency since 2002.
+
Build high-speed, data-intensive applications that elastically meet performance requirements at any scale.
+ Take advantage of Apache Geode's unique technology that blends advanced techniques for data replication, partitioning and distributed processing. + +

+ Apache Geode provides a database-like consistency model, reliable transaction processing and a shared-nothing architecture to maintain very low latency performance with high concurrency processing.

+ +

+ + Star + Fork + Watch + +

Replication and Partitioning

Data can easily be partitioned (sharded) or replicated between nodes allowing performance to scale as needed. Durability is ensured through redundant in-memory copies and disk-based persistence.

Persistence

Super fast write-ahead-logging (WAL) persistence with a shared-nothing architecture that is optimized for fast parallel recovery of nodes or an entire cluster.

Performance

Linear-scaling low latency for transactions, reads, writes and query processing of indexed or unindexed data.

In-Memory Storage

Blazing fast in-memory storage optimized for large heaps, with the option of using off-heap storage, compression and features such as disk-overflow, eviction and expiration of data.

Functions

Distributed location-aware user functions can be deployed and executed by the same nodes storing relevant sharded data for fast parallel processing. Failed operations can be retried on replicant nodes.

Transactions

ACID distributed transactions support efficient and safe coordinated operations on colocated data. Transactions can be initiated or suspended by either a client or a server.

OQL and Indexes

Object Query Language allows distributed query execution on hot and cold data, with SQL-like capabilities, including joins.
+ Multiple kinds of indexes can be defined and consistently maintained across the cluster.

Events

Clients can be notified about server-side data events, and servers can react synchronously or asynchronously with guaranteed delivery of ordered events.

Clustering

Highly scalable, robust advanced clustering technology with failure detection, dynamic scaling, and network-partition detection algorithms.

Performance is key. Consistency is a must.

Geode is an open source, distributed, in-memory database for scale-out applications.

Elastic performance

All data is stored in-memory for low latency. Performance scales linearly as nodes are added. Data is distributed automatically between nodes to optimize performance.

Database consistency

Data is persisted in write-optimized disk storage. Consistency checking is configurable between highest performance caching and ACID transactions.

Resilient clustering

Clusters fail-over to other nodes in case of failures, and rebalance remaining resources. Clusters self-heal when nodes return to operation.

Community

- To get in touch with other users and contributors of the Apache Geode (incubating) community: -

- Contributors: dev@geode.incubator.apache.org - (subscribe, - unsubscribe, archive) -
- JIRA notifications: issues@geode.incubator.apache.org - (subscribe, - unsubscribe, archive) -
- Commits: commits@geode.incubator.apache.org - (subscribe, unsubscribe, archive) -
- User questions at Stack Overflow -
- We use the Apache Geode JIRA for issue tracking -
- For other ways to contribute, see the Apache Geode (incubating) community page -

Contributors

- -

Development on the code base for Geode was started in 2002 as the commercial product GemFire. The open source version of Geode was released in April, 2015. Primary committers include members of the GemFire engineering team, customers, and ASF members.

- -

If you'd like to participate in Apache Geode (incubating), learn how to contribute.

Getting Started

- Read the quick start guide. -
- Check out the Getting Started page. -

Getting Started

Dive into How Tos

Other Resources

Training Materials