Scalability and best practices for Linux environment
As per your requirements there might be a need to monitor large number of systems, each with large number of metrics. To achieve this requirement, you need to run large number of scripts across the network.
The KM provides the ability to stage the scripts on dedicated systems where PATROL agents are installed. The KM monitors systems remotely, that is, from each PATROL Agent and manages execution of multiple scripts across multiple systems.
Consider the following points to plan about how to distribute the KMs work and load across multiple PATROL agents:
- How many systems will be remotely monitored by each PATROL Agent?
- How many scripts should be scheduled by each PATROL Agent?
- How many instances and parameters each PATROL Agent introduces?
To better plan such requirements, you can use the data in the following tables. The following setups are tested and reviewed as useful baseline for common use cases.
Each scenario is represented as a row in the table. The data provided here includes resource utilization of the PATROL Agent and the KMs backend JAVA collector.
Tested configuration scenarios:
| Load | Number of | Number of | Execution | Total instances created | Total | CPU load | Memory load (MB) |
|---|---|---|---|---|---|---|---|
| Small | 50 | 5 | 300 | 600 | 3,000 | 1 | 200 |
| Medium | 300 | 5 | 300 | 6000 | 40,000 | 8 | 800 |
| Large | 800 | 5 | 300 | 40,000 | 80,000 | 12 | 1,000 |
Comments
Log in or register to comment.