Overview - Amazon Redshift
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Overview

RA3 nodes provide the following advantages:

  • They are flexible to grow your compute capacity without increasing your storage costs. And they scale your storage without over-provisioning compute capacity.

  • They use high performance SSDs for your hot data and Amazon S3 for cold data. Thus they provide ease of use, cost-effective storage, and high query performance.

  • They use high bandwidth networking built on the Amazon Nitro System to further reduce the time taken for data to be offloaded to and retrieved from Amazon S3.

Consider choosing RA3 node types in these cases:

  • You need the flexibility to scale and pay for compute separate from storage.

  • You query a fraction of your total data.

  • Your data volume is growing rapidly or is expected to grow rapidly.

  • You want the flexibility to size the cluster based only on your performance needs.

To use RA3 node types, your Amazon Region must support RA3. For more information, see RA3 node type availability in Amazon Regions.

Important

You can use ra3.xlplus node types only with cluster version 1.0.21262 or later. You can view the version of an existing cluster with the Amazon Redshift console. For more information, see Determining the cluster maintenance version.

Make sure that you use the new Amazon Redshift console when working with RA3 node types.

In addition, to use RA3 node types with Amazon Redshift operations that use the maintenance track, the maintenance track value must be set to a cluster version that supports RA3. For more information about maintenance tracks, see Choosing cluster maintenance tracks.

Consider the following when using single-node RA3 node types.

  • Datasharing producers and consumers are supported.

  • To change node types, only classic resize is supported. Changing the node type with elastic resize or snapshot restore isn't supported. The following scenarios are supported:

    • Classic resize of a 1-node dc2.xlarge to a 1-node ra3.xlplus, and vice versa.

    • Classic resize of a 1-node dc2.xlarge to a multiple-node ra3.xlplus, and vice versa.

    • Classic resize of a multiple-node dc2.xlarge to a 1-node ra3.xlplus, and vice versa.

Working with Amazon Redshift managed storage

With Amazon Redshift managed storage, you can store and process all your data in Amazon Redshift while getting more flexibility to scale compute and storage capacity separately. You continue to ingest data with the COPY or INSERT command. To optimize performance and manage automatic data placement across tiers of storage, Amazon Redshift takes advantage of optimizations such as data block temperature, data block age, and workload patterns. When needed, Amazon Redshift scales storage automatically to Amazon S3 without requiring any manual action.

For information about storage costs, see Amazon Redshift pricing.

Managing RA3 node types

To take advantage of separating compute from storage, you can create or upgrade your cluster with the RA3 node type. To use the RA3 node types, create your clusters in a virtual private cloud (EC2-VPC).

To change the number of nodes of Amazon Redshift cluster with an RA3 node type, do one of the following:

  • Add or remove nodes with the elastic resize operation. In some situations, removing nodes from a RA3 cluster isn't allowed with elastic resize. For example, when a 2:1 node count upgrade puts the number of slices per node at 32. For more information, see Resizing clusters. If elastic resize isn't available, use classic resize.

  • Add or remove nodes with the classic resize operation. Choose this option when you are resizing to a configuration that isn't available through elastic resize. Elastic resize is quicker than classic resize. For more information, see Resizing clusters.

RA3 node type availability in Amazon Regions

The RA3 node types are available only in the following Amazon Regions:

  • US East (N. Virginia) Region (us-east-1)

  • US East (Ohio) Region (us-east-2)

  • US West (N. California) Region (us-west-1)

  • US West (Oregon) Region (us-west-2)

  • Africa (Cape Town) Region (af-south-1)

  • Asia Pacific (Hong Kong) Region (ap-east-1)

  • Asia Pacific (Hyderabad) Region (ap-south-2)

  • Asia Pacific (Jakarta) Region (ap-southeast-3)

  • Asia Pacific (Melbourne) Region (ap-southeast-4)

  • Asia Pacific (Mumbai) Region (ap-south-1)

  • Asia Pacific (Osaka) Region (ap-northeast-3)

  • Asia Pacific (Seoul) Region (ap-northeast-2)

  • Asia Pacific (Singapore) Region (ap-southeast-1)

  • Asia Pacific (Sydney) Region (ap-southeast-2)

  • Asia Pacific (Tokyo) Region (ap-northeast-1)

  • Canada (Central) Region (ca-central-1)

  • Canada West (Calgary) Region (ca-west-1)

  • China (Beijing) Region (cn-north-1)

  • China (Ningxia) Region (cn-northwest-1)

  • Europe (Frankfurt) Region (eu-central-1)

  • Europe (Zurich) Region (eu-central-2)

  • Europe (Ireland) Region (eu-west-1)

  • Europe (London) Region (eu-west-2)

  • Europe (Milan) Region (eu-south-1)

  • Europe (Spain) Region (eu-south-2)

  • Europe (Paris) Region (eu-west-3)

  • Europe (Stockholm) Region (eu-north-1)

  • Israel (Tel Aviv) Region (il-central-1)

  • Middle East (Bahrain) Region (me-south-1)

  • Middle East (UAE) Region (me-central-1)

  • South America (São Paulo) Region (sa-east-1)

  • Amazon GovCloud (US-East) (us-gov-east-1)

  • Amazon GovCloud (US-West) (us-gov-west-1)