Q55 — AWS SAA-C03 第3章

第 55/65 题 | ← 返回第3章

Q185.一家公司正在将应用程序从本地服务器迁移到 Amazon EC2 实例.作为迁移设计要求的一部分，解决方案架构师必须实施基础架构指标警报.如果 CPU 利用率在短时间内增加到 50% 以上，公司无需采取任何措施.但是，如果CPU使用率上升到50%以上，同时磁盘读取IOPS很高，公司就需要尽快行动了.解决方案架构师还必须减少误报.解决方案架构师应该怎么做才能满足这些要求?

正确答案: A. 尽可能创建 Amazon CloudWatch 复合警报.

解析

To address the company's migration design requirements effectively, let's analyze each option based on the stated criteria:Option A: Create Amazon CloudWatch composite alarms where possible.Explanation: Composite alarms allow you to combine multiple CloudWatch alarms into a single alarm, enabling more sophisticated logic. In this scenario, the company wants to be alerted only when both CPU utilization exceeds 50% and disk read IOPS are high simultaneously. Composite alarms can be configured to trigger based on the combination of these conditions, reducing the likelihood of false alarms that might occur if each condition were monitored separately.Relevance: This option directly addresses the need to act only when both conditions are met, making it the most suitable choice.Option B: Create Amazon CloudWatch dashboards to visualize the metrics and react to issues quickly.Explanation: Dashboards are useful for visualizing metrics and monitoring the overall health of your infrastructure. However, they do not provide automated alerting based on specific conditions. While dashboards can help in quickly identifying issues, they do not fulfill the requirement for automated action when specific conditions are met.Relevance: This option is more about monitoring and visualization rather than automated alerting.Option C: Create Amazon CloudWatch Synthetics canaries to monitor the application and raise an alarm.Explanation: Synthetics canaries are used to simulate user interactions with your application and monitor its availability and performance. They are more suited for monitoring the application's functionality and user experience rather than infrastructure metrics like CPU utilization and disk IOPS.Relevance: This option is not relevant to the specific requirements of monitoring infrastructure metrics.Option D: Create single Amazon CloudWatch metric alarms with multiple metric thresholds where possible.Explanation: While CloudWatch allows you to set multiple thresholds within a single alarm for a single metric, it does not support combining conditions across different metrics (e.g., CPU utilization and disk IOPS) in a single alarm. Therefore, this approach would not meet the requirement to trigger an alarm only when both conditions are met simultaneously.Relevance: This option is not feasible for the given requirements.Conclusion: The best approach to meet the company's requirements is to use Amazon CloudWatch composite alarms (Option A). Composite alarms allow you to combine multiple alarms into a single alarm, enabling you to set conditions that trigger an alert only when both CPU utilization exceeds 50% and disk read IOPS are high simultaneously. This reduces the likelihood of false alarms and ensures that the company takes action only when necessary.Answer: A. Create Amazon CloudWatch composite alarms where possible.为了有效地解决公司的迁移设计需求，让我们根据所述标准分析每个选项： 选项A：在可能的情况下创建Amazon CloudWatch复合告警。说明：复合警报允许您将多个CloudWatch警报组合成单个警报，从而实现更复杂的逻辑。在这种情况下，公司希望只有在CPU利用率超过50%且磁盘读IOPS同时很高时才会收到警报。可以将组合警报配置为基于这些条件的组合触发，从而减少如果单独监视每个条件可能发生的假警报的可能性。相关性：此选项直接解决了只有在两个条件都满足时才采取行动的需求，使其成为最合适的选择。选项B：创建Amazon CloudWatch仪表板，将指标可视化，并快速对问题做出反应。说明：仪表板对于可视化度量标准和监视基础结构的整体运行状况非常有用。但是，它们不提供基于特定条件的自动警报。虽然仪表板可以帮助快速识别问题，但当满足特定条件时，它们不能满足自动操作的要求。相关性：这个选项更多的是关于监控和可视化，而不是自动警报。选项C：创建Amazon CloudWatch合成金丝雀来监视应用程序并发出警报。解释：合成金丝雀用于模拟用户与应用程序的交互，并监视其可用性和性能。它们更适合于监控应用程序的功能和用户体验，而不是像CPU利用率和磁盘IOPS这样的基础设施指标。相关性：此选项与监视基础设施度量的特定需求无关。选项D：在可能的情况下，使用多个指标阈值创建单个Amazon CloudWatch指标警报。解释：虽然CloudWatch允许您在单个指标的单个告警中设置多个阈值，但它不支持在单个告警中组合不同指标（例如，CPU利用率和磁盘IOPS）的条件。因此，这种方法不能满足只有当两个条件同时满足时才触发报警的要求。相关性：该选项对于给定的需求是不可行的。结论:满足公司需求的最佳方法是使用Amazon CloudWatch复合告警（选项A）。复合告警可以将多个告警合并为一条告警，从而可以设置只有当CPU利用率超过50%且磁盘读IOPS同时很高时才触发告警的条件。这减少了误报的可能性，并确保公司只在必要时采取行动。 答：A.尽可能创建Amazon CloudWatch复合警报。