Memorystore for Redis Cluster is a fully managed Redis service for Google Cloud. Applications
running on Google Cloud can achieve extreme performance by leveraging the highly
scalable, available, secure Redis service without the burden of managing complex
Redis deployments.

## Key concepts and terms

### Hierarchical resource structure

Memorystore for Redis Cluster gathers the various resources used in a Redis deployment
into a hierarchical structure that simplifies administration and management.
Here is a diagram that illustrates this structure:

![A diagram that shows that a
Memorystore cluster is composed of a primary node and replica nodes](https://docs.cloud.google.com/static/memorystore/img/cluster-overview.png)

Memorystore for Redis Cluster instances are composed of a set of shards, each
containing a subset of your key space. Each shard in a Memorystore
cluster is composed of one primary node, and optionally up to five replica nodes.
When replica nodes are added, Memorystore automatically
distributes the nodes of a shard across zones to provide higher availability and
throughput.

### Instances

A Memorystore for Redis Cluster instance is a container for your data. The terms
*instance* and *cluster* can be used interchangeably when referring to a single
Memorystore for Redis Cluster unit of deployment. When provisioning a
Memorystore instance, you must provision enough shards to serve the
keyspace of your entire application.

For details about Memorystore cluster specification, see [Cluster and node specification](https://docs.cloud.google.com/memorystore/docs/cluster/cluster-node-specification.md.txt).

### Shards

Your cluster is composed of multiple equally sized shards. For details about
Memorystore shard specification, see [Cluster and node specification](https://docs.cloud.google.com/memorystore/docs/cluster/cluster-node-specification.md.txt).

### Primary and replica nodes

There's one primary node per shard. Each shard can have 0-5 replica
nodes. Replicas provide high availability and additional read throughput, and
are distributed across zones evenly.

For more details see [High availability and replicas](https://docs.cloud.google.com/memorystore/docs/cluster/ha-read-replicas).

### Redis version

Memorystore for Redis Cluster is based on open-source [Redis version 7.x](https://github.com/redis/redis/blob/7.0/00-RELEASENOTES) and supports a [subset of the
total Redis command library](https://docs.cloud.google.com/memorystore/docs/cluster/supported-commands).

## Cluster endpoints

Each instance has a discovery endpoint to which your client connects. Your
client also uses the discovery endpoint for cluster node discovery. For more
information, see [Cluster endpoints](https://docs.cloud.google.com/memorystore/docs/cluster/cluster-node-specification.md.txt#cluster_endpoints).

## Networking prerequisites

Before you can create a Memorystore for Redis Cluster instance, you must set up
[Networking](https://docs.cloud.google.com/memorystore/docs/cluster/networking.md.txt) for your project.

## Billing

See the [Pricing](https://docs.cloud.google.com/memorystore/docs/cluster/pricing) page for information on
pricing for available regions.