The performance of your Amazon RDS for MSSQL instance might be affected for multiple reasons, such as:
- Undersized hardware
- Changes in workload
- Increased traffic
- Memory issues
- Suboptimal query plans
Use a combination of these tools to identify the cause of slow-running queries:
- Performance Insights metrics (in RDS)
- Amazon CloudWatch metrics
- Enhanced Monitoring metrics
- Database statistics
- Native database tools
- CloudWatch metrics
To identify performance bottlenecks caused by insufficient resources, monitor these common CloudWatch metrics available for your Amazon RDS DB instance.
- CPUUtilization - Percentage of computer processing capacity used
- FreeableMemory - Available RAM on the DB instance
- SwapUsage - Swap space used by the DB instance
A higher percentage of CPU utilization generally indicates an active workload on the instance and the need for more CPU resources. Higher memory utilization along with swap space consumption indicates frequent swapping due to low memory availability for the workload. This might mean that your instance can't keep up with the workload. A high utilization of CPU and memory resources are typically caused by long-running queries, suddenly increased traffic, or a large number of idle connections.
If any of the above metrics are showing signs of peaking or exceeding their threshold, you may need to increase the size of the very small MSSQL instance to a larger one that is optimized for the performance area in question (CPU, memory, I/O throughput, network).
If this helps solve your problem, please choose this as the Accepted Answer so others on re:Post may benefit - thank you!
- Accepted Answerasked a year ago
- How do I move SQL Server users from one RDS for SQL Server instance to another or to an on-premises environment from RDS?AWS OFFICIALUpdated 9 months ago
- How can I identify what is blocking a query on a DB instance that is running Amazon RDS PostgreSQL or Aurora PostgreSQL?AWS OFFICIALUpdated 9 months ago
- AWS OFFICIALUpdated 2 years ago
- AWS OFFICIALUpdated 8 months ago
- EXPERTpublished 10 months ago
- EXPERTpublished 5 months ago