PROVIDER Β· Azure fully managed SQL database
Azure SQL Database
Azure SQL Database is the fully managed Azure relational service sharing the SQL Server engine, with a 99.995 percent SLA in Business Critical, Serverless auto-pause for intermittent load, and elastic pools for multi-tenant cost.
4.5pts
Pros
Cons
Product Information
Detailed Notes
Overview
Azure SQL Database started as SQL Azure in 2010 and is Microsoft's fully managed relational service on Azure sharing the SQL Server engine, available as single database, elastic pool, and managed instance with Hyperscale reaching hundreds of terabytes.
For Backend & Database work the advantage is operations-free hosting with SQL Server ecosystem parity, since no OS, instance, or backup management is needed, the Business Critical tier carries a 99.995 percent availability SLA, Serverless auto-pauses when idle, and elastic pools flatten cost across many tenant databases.
The limits are cost and shape, because vCore billing can exceed self-hosting under sustained load, elastic pools demand capacity planning, and some SQL Server features such as cross-database queries and SQL Agent are restricted in single database form, requiring managed instance for full parity.
Core Strengths
- The Business Critical tier carries a 99.995 percent SLA with built-in replicas and automatic failover
- Serverless auto-pauses when idle with per-second billing for intermittent workloads
- Hyperscale supports hundreds of terabytes with fast read-scale replicas
- Native Entra ID, Defender, and Power BI integration unifies identity and security
Product Ecosystem
Single database
The most common form with serverless and provisioned compute tiers, fitting standalone apps and per-tenant SaaS databases.
Elastic pool
Shares resources across many databases billed on aggregate usage, the key lever for multi-tenant SaaS cost.
Managed instance
Near instance-level SQL Server compatibility with cross-database queries and SQL Agent for lift-and-shift.
Drawbacks
- vCore billing can exceed self-hosted cost at sustained high load
- Single database lacks cross-database queries and SQL Agent without managed instance
- Poorly sized elastic pools cause contention and performance jitter
- Azure-bound, so cross-cloud migration rebuilds connectivity and security models
Use Cases
- Line-of-business primaries on Azure (β β β β β ) - Operations-free with SQL Server parity
- Multi-tenant SaaS shards (β β β β β ) - Elastic pools flatten cost
- Intermittent workloads and dev environments (β β β β ) - Serverless auto-pause
- Cloud-neutral deployment (β β ) - Strong platform coupling
Pricing
| Plan | Price | Scope |
|---|---|---|
| Serverless compute | Per vCore-second from about USD 0.000145 | Auto-pause with a 0.5 vCore minimum |
| Provisioned compute | Per vCore-hour | Better for steady load |
| Hyperscale | vCores plus storage replicas | 100 TB class with fast read replicas |
Azure SQL bills compute, storage, and backup in three parts with Serverless saving only when idle, elastic pools typically beating per-database provisioning by over 30 percent for many small tenant databases, and reserved capacity locking discounts under sustained full load.
FAQ
-
Azure SQL or SQL Server on a VM? Choose Azure SQL for operations-free hosting, fast elasticity, and high SLAs, and a VM when full instance features, special extensions, or extreme cost tuning matter.
-
When do elastic pools fit? Best for SaaS with dozens to thousands of small databases whose peaks interleave, while a handful of large databases contend inside one pool.
-
Single database or managed instance? Managed instance is required for unchanged applications needing cross-database queries or SQL Agent, while cloud-native apps usually start with single database plus pools.