Overview of Amazon Redshift Serverless and Pricing

Amazon Redshift is a fully managed, petabyte-scale data warehouse service in the cloud.  Amazon Redshift Serverless is a new feature of Amazon Redshift and was announced during Reinvent 2021.

What is Amazon Redshift Serverless ?

Amazon Redshift Serverless is a serverless option of Amazon Redshift that makes it easy to run and scale analytics in seconds without the need to set up and manage data warehouse infrastructure. With Redshift Serverless, any user—including data analysts, developers, business professionals, and data scientists—can get insights from data by simply loading and querying data in the data warehouse.

With the new serverless option, you can continue to query data in other AWS data stores, such as Amazon Simple Storage Service (Amazon S3) data lakes and Amazon Aurora and Amazon Relational Database Service (RDS) databases.

How is Amazon Redshift Serverless different from Amazon Redshift ?

Amazon Redshift is a leading Cloud data warehouse provider with 3x better price performance than other cloud data warehouses. Amazon Redshift is a managed service from AWS and though automated heavily, still involves management, configuration, tuning of the clusters to a large extent for customers with diverse use cases.

Redshift Serverless is introduced to address the following use cases

1. Focus on Insights, not on Managing the data warehouse – Access data without thinking about selecting instances, sizing, scaling, and tuning
2. Easy analytics – No need to create and manage clusters by selecting node types and node count
3. Variable Workloads – Scale up and down to meet the unknown, unpredictable demands of workload variations
4. Periodic Workloads – Automatic provisioning and scaling of resources so there is no manual intervention to create and manage compute
5. Steady state with spikes – Seamlessly and automatically scale up and down according to the workload spikes. Consistent performance at scale

Amazon Redshift Serverless Features

Following are some of the features of Redshift Serverless

1. Simplified user experience – Run and scale analytics without having to manage data warehouse clusters
2. Pay for use – Pay for the compute capacity only for the workload duration on a per-second basis.
3. Intelligent and dynamic compute – Automatically provisions and scales data warehouse capacity to deliver consistently fast performance.
4. All Redshift functionality and performance – Leverage Amazon Redshift’s rich SQL capabilities, seamless data lake integration, as well as industry-leading price performance at scale

Amazon Redshift Serverless Architecture

Following is high level Architecture of Amazon Redshift Serverless

As shown above, Amazon Redshift Serverless Integration with Storage, Access layer and automated features are similar to Amazon Redshift . Redshift Serverless has completely taken over the management, automation, tuning , configuration of the Compute and Acceleration layers , leaving Customer to focus on Data storage, access and analytics.

Migration from Amazon Redshift to Amazon Redshift Serverless

Migration to Amazon Redshift Serverless from Amazon Redshift is straight forward. No changes necessary for existing analytics and business intelligence applications. Simply point applications to the Amazon Redshift serverless end point and start running.

Redshift Serverless : Query your data lake

1. Query external data seamlessly from serverless end point.
2. Billed unified into RPU-hours – No per TB price for scanning data lake data.
3. Redshift Serverless routes data lake query through your VPC

Data sharing in Amazon Redshift Serverless

1. Share data seamlessly between provisioned cluster to serverless end point and vice versa
2. Share data across serverless end points across different AWS accounts

Amazon Redshift Spectrum Serverless

Amazon Redshift Spectrum is supported in Amazon Redshift Serverless. Setup and access of Redshift database objects stays same. Only difference is in pricing model. You no longer need to pay per TB scan cost for Redshift Spectrum queries, it is included with Amazon Redshift Serverless and charged by RPU usage

What is Redshift RPU?

1 Redshift RPU is 2 vCPU and 16 GB of memory

Amazon Redshift Serverless Pricing/Cost

Amazon Redshift Serverless Pricing has two components – Compute & Storage

Compute

1. Calculated based on RPUs. RPU is Redshift Processing Unit. Rate is $0.45 per RPU hour
2. You pay only for the compute capacity your data warehouse consumes when it is active.
3. Usage is billed in RPU-hours, metered on per-second basis
4. Base data warehouse, scaling capacity, data lake queries are part of same RPU-hours
5. Concurrency scaling and spectrum charges do not apply

Storage

1. Fixed GB-month rate pricing for the Redshift managed storage (RMS) and user snapshots. Rate is $0.024 per GB
2. Restore their data warehouse to specific points in last 24 hours at a 30 minute granularity at free of charge

Following is a sample Amazon Serverless Pricing Calculator Example

Assume you have loaded 100 GB of data into Redshift and one of your job will access this data daily for 2 hours. Lets say Amazon Redshift uses 100 RPUs capacity to run the job. Following is sample Daily Cost Calculator

Query period	2 hours (120 Minutes)
Capacity used	100 RPU
Daily Compute	$90 (2 x 100 RPU x $0.45 per RPU-hour)
Daily Storage	$12 (500×0.024)
Total Daily Cost	$112