Cloud Database Services and Managed DB Offerings

Cloud database services are hosted data management models in which the provider handles infrastructure provisioning, hardware maintenance, patching and scaling. They let organisations run relational and non-relational databases over the internet without maintaining on-premises servers.

Managed offerings are especially useful for workloads that need automatic backups, high availability and flexible scaling. They are widely used for transactional systems, analytics and modern application development.

Architecture and Operational Models

• Virtual machine deployment: Users rent cloud compute instances and install database software manually. They retain full administrative control over the operating system and configuration.
• Database-as-a-Service (DBaaS): The provider manages the database environment, including backups, replication and software upgrades.
• Serverless databases: Compute and storage scale dynamically, even down to zero, based on query load. Billing is tied to actual resource consumption.

Major Cloud Database Types

• Relational cloud databases: Designed for structured data with predefined schemas and Structured Query Language (SQL). Examples include Amazon Aurora, Google Cloud SQL and Azure SQL Database.
• NoSQL and document stores: Built for semi-structured or unstructured data such as key-value pairs, wide-column stores and JSON documents. Examples include MongoDB Atlas and Amazon DynamoDB.
• Cloud data warehouses: Optimised for analytical processing, business intelligence queries and aggregation of large datasets. Examples include Google BigQuery and Amazon Redshift.
• Vector databases: Specialised engines used to store high-dimensional embeddings for artificial intelligence similarity search and machine learning retrieval tasks.

Managed Database Platforms by Cloud Provider

Key Features and Operational Benefits

• Automated administration: Tasks such as patching, backups and upgrades are handled by the platform.
• Point-in-time recovery: Managed instances can be restored to an exact second before accidental deletion or corruption.
• Multi-region replication: Data can be replicated across regions to improve availability and business continuity.
• Failover support: During a regional outage, traffic can shift to a secondary data centre.
• Elastic scaling: Resources can be adjusted according to demand, which helps control cost and performance.
• Single-engine flexibility: Multi-model databases such as Azure Cosmos DB can support document, graph and key-value data structures within one system.

Exam-Relevant Points

• Public cloud spending: Cloud database services form one of the fastest-growing segments of public cloud infrastructure spending.
• Relational vs non-relational: Relational systems use structured schemas, while NoSQL systems are suited to flexible or semi-structured data.
• DBaaS model: The provider takes over core database operations, reducing the user’s administrative burden.
• Serverless model: Billing is based on actual usage rather than fixed provisioned capacity.
• Analytics focus: Data warehouses are used mainly for aggregation, reporting and business intelligence queries.
• AI use case: Vector databases support similarity search and retrieval in machine learning applications.
• Vendor lock-in: This is a key challenge when applications depend on proprietary extensions unique to one cloud provider.

Point-in-time recovery and automated multi-region replication are two high-value features frequently tested in prelims questions on managed databases.

Key Prelims Takeaways

• Cloud database services: Provider-managed data systems that remove the need for on-premises hardware.
• Virtual machine deployment: User-managed setup with manual installation of database software.
• DBaaS: Fully managed database service with automated administration.
• Serverless databases: Scale with demand and can scale down to zero.
• Amazon Aurora, Cloud SQL, Azure SQL Database: Important examples of managed relational databases.
• DynamoDB, Cosmos DB, Firestore: Key examples of managed NoSQL offerings.
• BigQuery and Redshift: Major cloud data warehouse platforms for analytics.