An educational assessment platform hosted on Google Cloud runs two primary microservices: a real-time Exam Submission API, where any request failure directly interrupts active student testing and breaches compliance rules, and an asynchronous Certificate Generation Service, which processes completion credentials within a 48-hour window. Currently, the engineering team applies a uniform 99.99% availability goal across all microservices, leading to excessive operational toil and unnecessary infrastructure cost. Which strategy should the Cloud Architect implement to properly align technical service levels with business impact?
- Define separate Service Level Objectives (SLOs) based on business criticality: establish a strict 99.99% availability SLO for the Exam Submission API, and define a lower 99.0% availability SLO for the asynchronous Certificate Generation Service to preserve error budget for core student workflows.Answer
- BDefine the target percentage goal of 99.0% as the Service Level Indicator (SLI) for the Certificate Generation Service, while setting its Service Level Objective (SLO) to equal the real-time ratio of successfully rendered certificates.
- CConfigure static metric threshold alerting on CPU utilization for both microservices, setting alerts to fire immediately whenever instance utilization exceeds 95% for more than 30 seconds.
- DConfigure backend load balancer health checks for the Certificate Generation Service to query the underlying database directly, removing background workers from the pool during transient database latency.
Answer
Establish tier-based Service Level Objectives (SLOs) aligned with business impact by assigning a strict 99.99% availability SLO to the critical synchronous Exam Submission API while adopting a relaxed 99.0% availability target for the asynchronous Certificate Generation Service.
Differentiating reliability targets according to business impact ensures operational resources and error budgets are focused where service disruption causes maximum business harm. High-impact synchronous services receive stringent SLOs, while background asynchronous services operate under relaxed targets.
Step-by-Step Solution
Key Concept
Aligning SLOs and Error Budgets with Business Impact