Serverless Analytics

Overview

Serverless Analytics provides real-time analytics on any data in SkySQL or cloud storage. No ETL (Extract, Transform, Load) is needed, since data is queried remotely.

Serverless Analytics allows ad-hoc high performance analytics without static provisioning. Pay only for what you use.

With SkySQL Serverless Analytics, you can:

  • Achieve predictable response time for concurrent users/apps

  • Join data across multiple repositories, such as S3 object storage, data warehouses, and SQL databases

  • Work with data in many formats, including JSON, CSV, Parquet, and AVRO

Serverless Analytics is powered by Apache Spark SQL.

Prerequisite

Serverless Analytics requires a compatible SkySQL service in the same region.

The compatible SkySQL service is used for authentication.

Compatible SkySQL services are:

  • Enterprise Server With Replica(s)

  • Xpand Distributed SQL

  • ColumnStore Data Warehouse

Launch Serverless Analytics

  1. Log in to the SkySQL portal.

  2. From the Dashboard, click the "+ Launch New Service" button.

  3. Choose the "Analytics" workload type.

  4. Choose the "Serverless Analytics" topology option. This topology is powered by Apache Spark SQL.

  5. Specify the other required options.

  6. Click the "Launch Service" button.

Connection Options

Serverless Analytics depends on a compatible SkySQL service.

To connect, Serverless Analytics uses the credentials for that service. For default credentials, see "Client Connections".

Connect with Notebook

Notebook enables interactive analytics with built-in visualization and forms for your Serverless Analytics service.

For details, see "Notebook".