A healthcare provider operates a telemedicine application on Google Cloud serving two distinct capabilities: a real-time emergency video triage service where disruptions create direct clinical risks and financial penalties, and an asynchronous medical record export service where generation can take up to 24 hours without violating compliance obligations. Currently, engineering applies a uniform 99.99% availability target across all infrastructure components and relies on static CPU utilization thresholds for alerting. During load surges, non-critical batch record export failures trigger high-priority paging alerts, consuming error budgets and exhausting on-call personnel. Which TWO architectural and operational adjustments should the lead cloud architect recommend to align technical service levels with business and clinical impact? (Select TWO.)
- Decouple service metrics by establishing separate SLIs and SLOs per user journey, maintaining a 99.99% availability SLO for real-time video triage while defining a relaxed completion latency SLO for asynchronous record exports.Cevap
- BAutomatically adjust the target SLO monthly to match the measured 30-day SLI historical baseline across all microservices, and contractually re-bind customer SLAs to this dynamic percentage.
- Transition alerting strategies from static CPU utilization thresholds to multi-window, multi-burn-rate alerts based on error budget consumption for critical user journeys.Cevap
- DConfigure Cloud Load Balancing health checks for both services to execute deep dependency queries against backend databases to trigger immediate multi-region failover when backend CPU spikes occur.