Azure SQL Database Compute Options
Selecting the right compute option for your Azure SQL Database is crucial for balancing performance, cost, and scalability. Azure SQL Database offers a variety of compute models designed to meet diverse application needs.
Understanding Compute Models
Azure SQL Database provides two primary compute models:
- DTU (Database Transaction Unit): A bundled measure of compute, memory, and I/O resources. It's a simple, unified measure that is easy to understand and predict.
- vCore (Virtual Core): Offers more flexibility and control over your resources. This model allows you to scale compute and storage independently and choose hardware generations.
DTU Model
The DTU model is suitable for applications with predictable workloads. It simplifies resource management by combining compute, memory, and read/write I/O into a single metric.
Service Tiers (DTU Model)
- Basic: Ideal for development, testing, and small, low-traffic workloads.
- Standard: A good choice for most common applications requiring moderate performance.
- Premium: Designed for mission-critical applications requiring high performance and availability, with features like automated backups and geo-replication.
Each tier offers different performance levels represented by DTU counts (e.g., 5 DTUs, 10 DTUs, 20 DTUs, etc.).
vCore Model
The vCore model provides greater transparency and control. You can choose your compute tier, hardware generation, and number of vCores, and scale compute and storage independently.
Service Tiers (vCore Model)
- General Purpose: Balances price and performance for most business workloads. Offers up to 99.99% availability.
- Business Critical: Provides the highest performance and availability for mission-critical workloads. Features include in-memory OLTP, read-scale availability, and automated backups. Offers up to 99.995% availability.
- Hyperscale: Designed for very large databases (up to 100 TB) and high-throughput workloads, offering elastic scaling of compute and storage.
Hardware Configurations
Within each service tier (except Hyperscale, which has its own scaling), you can select hardware generations (e.g., Gen5) and specify the number of virtual cores and memory. This allows for fine-grained resource allocation.
Choosing the Right Option
Consider the following factors when making your decision:
- Workload Predictability: If your workload is stable and predictable, the DTU model might be simpler. For variable or complex workloads, vCore offers more flexibility.
- Resource Control: The vCore model gives you direct control over CPU, memory, and I/O.
- Scalability Needs: If you anticipate rapid growth or need to scale compute and storage independently, vCore, particularly Hyperscale, is advantageous.
- Performance Requirements: For the most demanding applications, Business Critical tiers in the vCore model offer the highest performance.
- Cost Optimization: Both models offer different price points. Evaluate your usage patterns to find the most cost-effective solution.
Recommendation: For new applications, the vCore model is generally recommended due to its flexibility and control. However, the DTU model remains a valid and often simpler choice for existing applications or those with well-understood, stable resource needs.
To get started with configuring your Azure SQL Database compute options, visit the Azure portal or consult the official Azure SQL Database documentation.
Explore Pricing Learn More (vCore)