Capacity Planning
Capacity planning provides the analytical data for a group of monitors, and is used to determine the capacity required to optimize resources for a particular operation or workload. This allows you to plan, process, and procure additional resources based on the metrics-related data collected.
Table of contents
Why do you need capacity planning?
Capacity planning helps you manage the capacity of the infrastructure to ensure optimal performance for the workloads.
With capacity planning you can:
- View and track the capacity utilization of your resources based on metrics like CPU, memory, and storage.
- Monitor the health or status of your resources grouped under the capacity monitor.
- Forecast and manage the capacity of your resources.
- Coordinate the workload among resources with Top-N/Bottom-N Analysis.
- Find the outliers, the data points beyond the average value, and maintain the quality of your resources.
- Diagnose service issues in your resources and drill down to the root cause of the issue.
Use case
- When you run multiple resources to perform one business application, finding the under/over-utilized resources is tough. It is stressful to monitor each resource individually, collect their performance stats, and record their behavior. By integrating those resources into capacity planning, you can determine the required capacity for the infrastructure. With capacity planning, you can seamlessly track the size, space, network, and capability of resources, irrespective of the number.
- Manage variable resources and optimize your application performance
Capacity Planning also allows you to monitor the behavior of auto-scaling monitors based on the aggregated data collected for several instances.
For example, say there is an instance tagged as env:zylker and a Capacity Planning monitor created for it as zylker. If an ip-123 instance with env:zylker tag gets launched at 10am, it gets aggregated in the zylker capacity monitor. Suppose the ip-123 instance with env:zylker tag is killed at 11am and another instance copy ip-456 gets launched at 11:01am with the same tag. The data from the new instance gets aggregated into the same Capacity Planning monitor: zylker.
Using this method, you can maintain data for the required instance. Note that the tag definition should be maintained accurately. Tags can be defined on the Site24x7 console as well, but it’s best done on the AWS side for appropriate mapping during discovery. You can also obtain details like the number of times an instance gets spawned and killed.
Licensing
Capacity planning is available for all licensed users. You can add up to 50 capacity planning monitors for each Site24x7 account. For MSP accounts, you can add up to 50 monitors for each customer or Business Unit.