Which SLIs/SLOs are relevant for Ingress vs Gateway API?

In the context of cloud-native applications, both Ingress and Gateway APIs serve critical roles in managing traffic flow. However, the key SLIs (Service Level Indicators) and SLOs (Service Level Objectives) that are relevant may vary based on the configuration and specific use cases. Below are some important SLIs and SLOs relevant to Ingress and Gateway API:

Key SLIs and SLOs

• Response Time: Measure the time taken to serve requests through Ingress or Gateway API. Response time should be less than 200ms for 95% of requests.
• Error Rate: Monitor the number of failed requests relative to total requests. Error rate should be lower than 1%.
• Availability: Track the uptime of services routed through Ingress or Gateway API. Service availability should be above 99.9%.
• Throughput: Measure the number of requests processed per second. Throughput should meet the scaling requirements for user traffic.
• Latency: Understand the time taken for a request to travel from the client to the server and back. Keep latency under 100ms for 95% of requests.

Monitoring and Observability

Utilizing tools such as Prometheus and Grafana can help in continuously monitoring these SLIs and ensuring that SLOs are met, thus maintaining the desired service quality.