How to interpret my Compute Metrics?

General Information

For general information on monitoring, please refer to the dedicated article (How do I monitor my platforms performance?).

Getting Started

Compute metrics provide meaningful information about compute capabilities of the platforms web servers.

Available Metrics

CPU Usage (%)

• CPU IO Wait
  The percentage of time that the CPU is waiting for input/output operations aggregated across all web servers. A high CPU IO Wait may indicate that the database is under-provisioned, meaning that it is not meeting the performance requirements of your workload.
• CPU System
  The percentage of time that the CPU is handling system operations aggregated across all web servers.
• CPU Application
  The percentage of time that the CPU is handling user operations aggregated across all web servers. A high CPU Application may indicate that:
  • your source code is consuming a lot of CPU and you should investigate and optimize it for performance
  • the platform is under-provisioned and does not meet the performance requirements of your workload

Memory Usage (Mo)

• Memory Used
  The amount of memory in use aggregated across all web servers. A high memory usage may indicate that:
  • your source code is consuming a lot of RAM and you should investigate and fix issues (memory leaks, etc.)
  • the platform is under-provisioned and does not meet the performance requirements of your workload
• Memory Swap
  The amount of swap space aggregated across all web servers. When happening, contents not currently in use are swapped to a disk to make the memory available for other applications or processes. A high memory swap can severely slow down the performance of your application.

Load (must be under 1.0)

The load averaged over a 5 minute window.