Constellation ShortList™ Automated Cloud Data Warehouse Services

Analytical relational database management systems (RDBMSs) have emerged as one of the most popular types of database services offered on public clouds. More popularly known as cloud data warehouse services, these offerings simplify the deployment and ongoing use of data warehouses on a small, moderate, or massive scale.

Any cloud service eliminates software and hardware deployment tasks and eases administrative burdens, but the standouts among cloud data warehouse services offer more automated capabilities. They take advantage of cloud-native automation capabilities, such as serverless scaling and elastic compute and storage provisioning that grow or shrink dynamically, based on usage. In addition, highly automated cloud data warehouse services add automated self-tuning and query optimization features that minimize, if not eliminate, previously manual administrative and systems-management tasks. Automated capabilities enable teams to focus on harnessing new data sources and supporting new workloads.

This ShortList presents Constellation’s pick of highly automated cloud data warehouse services. The vendors included on this ShorList is unchanged since its 2023 publication.

Threshold Criteria

Constellation considers the following criteria for these solutions:

• Automated cloud data warehouse services enable customers to set up database instances through simple interfaces without having to make complex decisions or fixed commitments on storage or compute infrastructure.
• The service should incorporate artificial intelligence/ machine learning-based augmented optimization features and plans for or previews of generative AI-based capabilities.
• The service automatically scales by using serverless cloud compute capacity, sizing to initial requirements and scaling up and down as data stores and workloads change.
• Database management is also automated and responds to changes in scale and workloads, with no manual indexing or partitioning decisions to be made.
• The service is self-tuning, continuously learning from and optimizing the performance of workloads and queries rather than simply enforcing user-defined rules.
• Even automated products need some guidance from humans, so there should be provisions for setting workload priorities and service levels.
• Products should provide visibility into how automated performance-tuning decisions are made as well as ways to both boost performance and control costs.
• Since these are managed services, customers should not have to worry about software patches, upgrades, or backups.