What cloud-based monitoring service should I use for CPU, RAM, disk, and syslog?
Asked by
MellowCedar42
I currently use New Relic for part of my infrastructure and a self-hosted Grafana plus Prometheus setup for the rest. I would like to move away from maintaining Grafana and use a managed cloud service instead. My main requirements are monitoring CPU, memory, and disk usage, with syslog error collection as an optional extra. New Relic is more expensive than I would like, so I'm looking for affordable alternatives or recommendations. Ideally, I'd like to compare the new service with my existing setup before moving everything over.
Answers
If your priority is getting rid of the self-hosted stack, I’d compare managed services such as Site24x7, Honeybadger, or similar platforms. They can handle server metrics and basic log or syslog monitoring without requiring you to operate Prometheus and Grafana. Make sure the plans include the number of hosts, log retention, alerting, and any custom metrics you need.
Reply to this answer
Managed Prometheus is probably the least disruptive option. You can keep your existing exporters and alert rules while moving storage and operations to a provider. Compare retention limits, metric-cardinality pricing, and especially log-ingestion costs. CPU, memory, and disk metrics are usually inexpensive, but syslog volume can make the bill grow quickly. I’d also keep a small external heartbeat monitor so the monitoring system itself isn’t in the same failure domain as the infrastructure it watches.
Reply to this answer
Before choosing a product, clarify whether syslog needs full search and analysis or whether you only need alerts for a few error patterns. For basic server metrics, almost any agent-based monitoring service will work. Full log management is usually the feature that separates the pricing tiers, so treating metrics and logs as separate requirements may give you a cheaper and more flexible setup.
Reply to this answer
Since you want to run the new service alongside your current setup first, look for a platform with an easy agent installation, straightforward export or API access, and a usable free trial. Several newer OpenTelemetry-based services position themselves as lower-cost replacements for New Relic and Grafana/Prometheus, but verify how they price logs and long-term retention before committing.
Reply to this answer
Checkmk is worth a look if you want an all-in-one platform with discovery, dashboards, alerting, and server monitoring. It’s easy to start with the free self-hosted version, and there are managed options if you later decide not to maintain the system yourself. Netdata and Beszel are also simple choices for basic CPU, RAM, and disk visibility.
Reply to this answer
Add your answer
Answers are reviewed before publication.
Related Discussions