Kubernetes and Cloud Native Associate (KCNA)Cloud Native ObservabilityMedium
A team is managing a Kubernetes cluster with several stateless microservices. They want to set up an effective monitoring system that adheres to the 'four golden signals' of monitoring: Latency, Traffic, Errors, and Saturation. For a typical HTTP-based microservice, which metric would best represent 'Traffic'?
- AMemory usage in MB
- BNumber of active network connections
- CCPU utilization percentage
- DRequests per second (RPS)
Show answer & explanationAnswer & explanation
Correct answer: D. Requests per second (RPS)
For an HTTP-based microservice, 'Traffic' is best represented by the rate of requests it receives, typically measured as Requests Per Second (RPS). This metric directly indicates the demand on the service. CPU utilization and memory usage relate to 'Saturation', and active network connections are a related but less direct measure of the workload 'Traffic' for an HTTP service.
Why the other options are wrong
- A. Memory usage is a 'Saturation' metric, indicating resource consumption.
- B. Number of active network connections can contribute to saturation but does not directly quantify the 'Traffic' (workload) for an HTTP service.
- C. CPU utilization is a 'Saturation' metric, indicating resource consumption.
Four Golden Signals (Traffic)
One of the 'four golden signals' of monitoring, 'Traffic' measures the demand being placed on a system, typically quantified as requests per second or throughput.
- Indicates the amount of work a service is doing.
- Essential for understanding service load and scaling needs.
- Often measured as RPS for web services or bytes/second for network services.
Memory trick: LET'S monitor: Latency, Errors, Traffic, Saturation.