Check Neptune CPU, request rate, memory and storage in WHMCS
As the hosting operator, use PD AWS Neptune’s customer graphs alongside your team’s application checks. CPU, request rate, freeable memory and allocated storage describe different pressures. Read their labels and time range before deciding which investigation is needed.
Customer steps: read the service graphs
- Open the active service and confirm its Neptune cluster and AWS region.
- Choose Usage Graphs where your hosting business makes that section available.
- Review CPU utilisation, Requests per second, Freeable memory and Storage used.
- Compare the displayed period with application slowdowns and recent Activity.
- Keep the affected period and visible measurements before changing the service.
The charts request recent CloudWatch data when opened, covering the previous 24 hours. They are not continuous live monitoring or measurements for one graph query. CPU is a percentage, memory and storage use their displayed scales, and requests per second is a rate rather than a connection count.
Interpret the graph correctly
The request-rate graph covers server requests from all sources. It is not proof of successful application queries, query latency or one application’s traffic. Compare it with the application’s own observations rather than treating every increase as a fault.
The Storage used graph reports a volume-allocation measure used for storage billing, not just the current size of graph data or remaining free capacity. Do not infer that removing graph data will immediately reduce the displayed value or provider charges. Your hosting team should verify actual capacity and provider billing separately when those are the concern.
If data is missing or pressure continues
A new, stopped or unavailable cluster may lack a useful recent series. Provider reporting behaviour, read failures and monitoring permissions can also affect the display. Blank graphs do not establish zero load or unlimited capacity. Your hosting team should compare the intended resource and provider read before changing the plan.
CPU pressure, query demand and memory pressure require different checks. A larger instance does not itself establish that a slow-query cause is resolved. If graphs load correctly but queries remain slow, continue workload diagnosis with your database/application administrator.
See the PD AWS Neptune WHMCS guide for related tasks, or explore PD AWS Neptune.
