An enterprise AI application experiences intermittent HTTP 429 (Too Many Requests) errors when calling Azure OpenAI Service during peak hours. You need to analyze the exact token consumption trends and throttling patterns over the past week. Where should you look first in Azure Monitor?
Trap 1: Log Analytics workspace querying the AzureActivity table for…
Incorrect. AzureActivity logs management plane operations, whereas data plane usage metrics and diagnostic logs track inference token counts.
Trap 2: Azure Cost Management and Billing dashboard for daily spend trends
Incorrect. Cost Management tracks monetary expenditure, not real-time API throttling or token throughput.
Trap 3: Application Insights availability tests configured with a 5-minute…
Incorrect. Availability tests ping a URL to check uptime, but do not provide granular Azure OpenAI token usage metrics.
- A
Azure Monitor Metrics explorer, charting 'Processed Tokens' and 'Token Generation' metrics filtered by deployment name
Metrics explorer exposes the Processed Tokens and Token Generation metrics per deployment, revealing consumption trends and throttling correlation over the past week. This directly satisfies the need to analyse exact token consumption and 429 patterns without querying logs.
- B
Log Analytics workspace querying the AzureActivity table for throttling events
Why it fails: Incorrect. AzureActivity logs management plane operations, whereas data plane usage metrics and diagnostic logs track inference token counts.
- C
Azure Cost Management and Billing dashboard for daily spend trends
Why it fails: Incorrect. Cost Management tracks monetary expenditure, not real-time API throttling or token throughput.
- D
Application Insights availability tests configured with a 5-minute frequency
Why it fails: Incorrect. Availability tests ping a URL to check uptime, but do not provide granular Azure OpenAI token usage metrics.