Redshift serverless is a fully managed database service for Amazon Redshift that uses the AWS Lambda serverless computing platform. It enables you to run SQL queries on data stored in Amazon S3 and other sources by executing them from within your application code. With Redshift serverless, you get an easy-to-use web application that lets you define the schema of your tables and then run queries against them.
Serverless is a term floating around cloud computing for a while now, but what exactly is it? Serverless is a system where you don't have to manage servers or deal with scaling issues. Instead, you can focus on building your application and let the providers handle everything else.
Spectrum is an extension of AWS Lambda that allows you to run serverless applications on multiple clouds at once (currently including Google Cloud Platform). With Spectrum, you can write code using familiar languages like Python, Java, and NodeJS, then run them across multiple platforms without worrying about portability issues or vendor lock-in.
There are no dedicated resources on Redshift serverless, so you don't need to purchase or manage hardware. You also don't have to worry about hardware failure or upgrades, maintenance, security, and other issues with owning infrastructure.
You can focus on your application rather than managing servers.
With Redshift serverless, you can scale storage independently of compute. This means that if you want to add more capacity to your database but don't need more processing power, you can do it without having to provide any new computer.
The same holds true in reverse: you can scale compute independently of storage by scaling up or down simultaneously. This makes Redshift serverless especially useful for workloads with unpredictable traffic patterns and cyclical spikes in usage.
Redshift Serverless automatically scales storage based on the size of your data and can automatically scale up or down as your data grows or shrinks.
When you create a cluster, it comes with a set amount of disk space from Amazon S3. The minimum disk size is 5 GB, but you can increase that limit by specifying an Amazon S3 bucket name for your cluster in the Redshift Serverless console (e.g., mybucket). The default bucket name is "redshift-cluster-data," which will also be used if you don't specify anything else during setup.
You can manually scale up/down your storage by clicking on the Storage tab in the left navigation bar in the console.
The first thing to know about Redshift is that AWS fully manages it.
This means you don't have to worry about managing your data warehouse's physical hardware, storage, or security. Instead, you can focus on building your data pipeline and querying your data using SQL-like queries called Amazon Redshift Spectrum Query Language (ASQL).
The second thing to know about Redshift is that it uses shared-everything architecture. This means that all nodes in a cluster share storage and compute power, but each node has its database engine process (known as an instance).
As a user, you pay for what you use. You can scale storage and compute independently. You can scale up or down in real-time. In fact, as a serverless user, there are only two things that are charged: data transfer and data storage.
Redshift Serverless supports external tables, which means you can use the same Redshift data in a SQL query using traditional SQL. In addition to that, Redshift Serverless also supports UDFs. The usage of UDFs is identical to regular UDFs. You can call them as part of your query, and they will be executed as part of the query plan at runtime.
Redshift serverless is the future. It's a great choice for your data warehouse because it gives you all the power of Redshift but with much less complexity. The only thing you need to manage is an AWS account and an API key or password, and everything else runs itself in the background!
Future proof your next project using Redshift serverless—you'll be glad you did!
Redshift serverless is the future of data warehousing. In fact, it's already here. Redshift serverless allows you to build a data warehouse without having to manage infrastructure or worry about scaling. It brings together all of the best features from other databases and makes them available in one place — with no upfront commitment!