Azure Relational Databases#
The Relational Database Building Block deploys a relational database through the Azure Database services. It provides cost-efficient and resizable capacity while automating time-consuming administration tasks such as hardware provisioning, database setup, patching and backups.
Interfaces#
- Programmatic SDKs: each database type has SDKs available in multiple programming languages
- CLI: each database type can also be accessed from the CLI, if the necessaryh packages are installed
- API: each database type can also be accessed from an API, e.g. ODBC
Configurations#
The user must choose the type of relational database.
Microsoft SQL Server#
Microsoft SQL Server is a widely-used proprietary relational database from Microsoft that integrates. It is an integrated transaction safe, ACID-compliant database with full commit, rollback, crash recovery, and row-level locking capabilities. Due to it being a proprietary solution of Microsoft, it integrates seasmlessly with Azure and offers additional capabilities compared to the open source variants.
Hardware configuration#
To create a Microsoft SQL Server database, you need to select the service tier and a number of Database Transaction Units (DTUs). A DTU represents a blended measure of CPU, memory, reads, and writes; which makes it easier to compare the relative performance between the different service tiers. During a technical meeting, more clarifications will be provided regarding the benchmarks that Microsoft performs around these DTUs, and how many DTUs would fit your use-case.
The logical CPUs cannot be chosen, but are either Intel E5-2673 v3 (Haswell), Intel E5-2673 v4 (Broadwell) or Intel SP8160 (Skylake)-based CPUs.
The following tables describes the limits and characteristics of the different service tiers:
| Database Tier | Basic | Standard | Premium |
|---|---|---|---|
| Uptime SLA | 99.99% | 99.99% | 99.99% |
| Maximum backup retention | 7 days | 35 days | 35 days |
| CPU | Low | Low, Medium, High | Medium, High |
| IO throughput (approximate) | 1-5 IOPS per DTU | 1-5 IOPS per DTU | 25 IOPS per DTU |
| IO latency (approximate) | 5 ms (read), 10 ms (write) | 5 ms (read), 10 ms (write) | 2 ms (read/write) |
| Columnstore indexing | N/A | S3 and above | Supported |
| In-memory OLTP | N/A | N/A | Supported |
| Maximum storage size | 2 GB | 1 TB | 4 TB |
| Maximum DTUs | 5 | 3000 | 4000 |
MySQL#
MySQL is the world’s most popular open source database, enabling the cost-effective delivery of reliable, high-performance and scalable Web-based and embedded database applications. It is an integrated transaction safe, ACID-compliant database with full commit, rollback, crash recovery, and row-level locking capabilities.
Hardware configuration#
Different database tiers and compute resources are available for the MySQL database type.
All logical CPUs are based on Intel E5-2673 v4 (Broadwell) 2.3-GHz processors.
CPU and memory are bound by a fixed ratio (see Memory per vCore). Thus to increase your memory within a database tier, you also need to scale CPU (the number of vCores) and vice versa.
Additionally, the IOPS are bound by a fixed ratio to the storage (3 IOPS/GB for General Purpose and Memory Optimized). Thus to increase the IOPS, you also need to scale your storage size.
The following table describes the supported sizes & tiers:
| Database Tier | Basic | General Purpose | Memory Optimized |
|---|---|---|---|
| vCores | 1, 2 | 2, 4, 8, 16, 32, 64 | 2, 4, 8, 16, 32 |
| Memory per vCore | 2 GB | 5 GB | 10 GB |
| Storage size | 5 GB to 1 TB | 5 GB to 16 TB | 5 GB to 16 TB |
| Storage increment size | 1 GB | 1 GB | 1 GB |
| IOPS | Variable | 3 IOPS/GB Min 100 IOPS Max 20,000 IOPS |
3 IOPS/GB Min 100 IOPS Max 20,000 IOPS |
| Database backup retention period | 7 to 35 days | 7 to 35 days | 7 to 35 days |
Postgres#
PostgreSQL has become the preferred open source relational database for many enterprise developers and start-ups, powering leading business and mobile applications.
Hardware configuration#
Different database tiers and compute resources are available for the Postgres database type.
All logical CPUs are based on Intel E5-2673 v4 (Broadwell) 2.3-GHz processors.
CPU and memory are bound by a fixed ratio (see Memory per vCore). Thus to increase your memory within a database tier, you also need to scale CPU (the number of vCores) and vice versa.
Additionally, the IOPS are bound by a fixed ratio to the storage (3 IOPS/GB for General Purpose and Memory Optimized). Thus to increase the IOPS, you also need to scale your storage size.
The following table describes the supported sizes & tiers:
| Database Tier | Basic | General Purpose | Memory Optimized |
|---|---|---|---|
| vCores | 1, 2 | 2, 4, 8, 16, 32, 64 | 2, 4, 8, 16, 32 |
| Memory per vCore | 2 GB | 5 GB | 10 GB |
| Storage size | 5 GB to 1 TB | 5 GB to 16 TB | 5 GB to 16 TB |
| Storage increment size | 1 GB | 1 GB | 1 GB |
| IOPS | Variable | 3 IOPS/GB Min 100 IOPS Max 20,000 IOPS |
3 IOPS/GB Min 100 IOPS Max 20,000 IOPS |
| Database backup retention period | 7 to 35 days | 7 to 35 days | 7 to 35 days |
Database tiers and sizes#
Microsoft SQL Server#
To create a Microsoft SQL Server database, you need to select the service tier and a number of Database Transaction Units (DTUs). A DTU represents a blended measure of CPU, memory, reads, and writes; which makes it easier to compare the relative performance between the different service tiers. During a technical meeting, more clarifications will be provided regarding the benchmarks that Microsoft performs around these DTUs, and how many DTUs would fit your use-case.
The logical CPUs cannot be chosen, but are either Intel E5-2673 v3 (Haswell), Intel E5-2673 v4 (Broadwell) or Intel SP8160 (Skylake)-based CPUs.
The following tables describes the limits and characteristics of the different service tiers:
| Database Tier | Basic | Standard | Premium |
|---|---|---|---|
| Uptime SLA | 99.99% | 99.99% | 99.99% |
| Maximum backup retention | 7 days | 35 days | 35 days |
| CPU | Low | Low, Medium, High | Medium, High |
| IO throughput (approximate) | 1-5 IOPS per DTU | 1-5 IOPS per DTU | 25 IOPS per DTU |
| IO latency (approximate) | 5 ms (read), 10 ms (write) | 5 ms (read), 10 ms (write) | 2 ms (read/write) |
| Columnstore indexing | N/A | S3 and above | Supported |
| In-memory OLTP | N/A | N/A | Supported |
| Maximum storage size | 2 GB | 1 TB | 4 TB |
| Maximum DTUs | 5 | 3000 | 4000 |