Spring Cloud Research & Evidence-Based Practice 3

Question 1

Case studies of teams migrating monoliths to Spring Cloud microservices most frequently cite which anti-pattern as the root cause of performance regression?

Accepted Answer

Distributed monolith: microservices that require many synchronous calls to peers for every request, creating chatty inter-service communication

Answer

A 'distributed monolith' retains tight coupling through excessive synchronous inter-service calls, negating microservices benefits and adding network overhead compared to the original monolith.

Question 2

Evidence-based guidance on Resilience4j circuit breaker sliding window sizing recommends which approach when establishing initial thresholds?

Accepted Answer

Set window size to 100 and failure threshold to 50% based on production traffic baselines, then refine using real error-rate data

Answer

Starting from observed production baselines (traffic volume, normal error rates) and iteratively refining thresholds prevents both false-positive trips and missed failures.

Question 3

When reviewing Micrometer metrics emitted by Spring Cloud applications in Grafana, which metric is most useful for detecting upstream service degradation before users are impacted?

Accepted Answer

http.client.requests p99 latency trending upward over time

Answer

Rising p99 latency on outbound HTTP client calls indicates a deteriorating upstream service, providing a leading indicator before error rates spike or circuit breakers open.

Question 4

A research review comparing Spring Cloud Kubernetes versus Spring Cloud Netflix for service discovery finds which primary operational advantage of Spring Cloud Kubernetes?

Accepted Answer

It leverages native Kubernetes Services and ConfigMaps, eliminating the need to run and maintain a separate Eureka server

Answer

Spring Cloud Kubernetes uses the Kubernetes control plane for service discovery and configuration, removing the operational burden of running Eureka as additional infrastructure.

Question 5

Evidence from SRE practice suggests that a Spring Cloud LoadBalancer health-check interval should be determined by which criterion?

Accepted Answer

The maximum time an unhealthy instance can serve traffic before causing a measurable SLO violation

Answer

Health-check interval should be driven by how quickly an unhealthy instance must be removed to keep errors within SLO bounds — working backward from the error budget.

Spring Cloud Practice Test

Spring Cloud Practice Test

Spring Cloud Research & Evidence-Based Practice 3