A company hosts a static website on Amazon S3 and provides APIs via Amazon API Gateway that invoke Lambda functions. During spikes in traffic, users report intermittent slowness (no failed requests). A developer needs to identify slow executions across all Lambda functions. Which approach meets this need?

Enabling AWS X-Ray for all Lambda functions and using X-Ray insights is the most effective solution because X-Ray is designed for end-to-end tracing and performance analysis of distributed applications. It provides detailed timing information for each segment of a request, including Lambda function execution, allowing you to pinpoint bottlenecks and identify slow executions across all integrated services. X-Ray Insights specifically highlight anomalies and performance deviations. Querying CloudWatch Logs Insights is useful for log analysis but requires parsing and aggregating execution durations from logs, which is less efficient and comprehensive for tracing than X-Ray. CloudTrail Insights focuses on API activity anomalies, not application performance or Lambda execution duration. Using AWS Glue for log processing is overly complex and introduces significant latency for real-time performance analysis compared to X-Ray's integrated tracing capabilities.