4.5 Lab: Monitoring GenAI

aws bedrock put-model-invocation-logging-configuration \
  --logging-config '{
    "cloudWatchConfig": {
      "logGroupName": "/aws/bedrock/model-invocations",
      "roleArn": "<YOUR_LOGGING_ROLE_ARN>",
      "largeDataDeliveryS3Config": {
        "bucketName": "<YOUR_S3_BUCKET>",
        "keyPrefix": "bedrock-logs/"
      }
    },
    "textDataDeliveryEnabled": true,
    "imageDataDeliveryEnabled": false,
    "embeddingDataDeliveryEnabled": false
  }'

import boto3
import json

client = boto3.client('bedrock-runtime', region_name='us-east-1')

prompts = [
    "What is Amazon Bedrock?",
    "Explain RAG in simple terms.",
    "List 3 benefits of serverless computing.",
    "What is prompt engineering?",
    "How does vector search work?"
]

for prompt in prompts:
    response = client.invoke_model(
        modelId='anthropic.claude-3-5-sonnet-20241022-v2:0',
        contentType='application/json',
        accept='application/json',
        body=json.dumps({
            "anthropic_version": "bedrock-2023-05-31",
            "max_tokens": 128,
            "messages": [{"role": "user", "content": prompt}]
        })
    )
    result = json.loads(response['body'].read())
    print(f"Prompt: {prompt[:40]}... | Tokens: input={result['usage']['input_tokens']}, output={result['usage']['output_tokens']}")

fields @timestamp, modelId, inputTokenCount, outputTokenCount
| sort @timestamp desc
| limit 20

fields @timestamp
| stats count() as invocations by bin(1h)

fields @timestamp, inputTokenCount, outputTokenCount
| stats sum(inputTokenCount) as totalInput, sum(outputTokenCount) as totalOutput by bin(1h)

fields @timestamp, @duration
| stats avg(@duration) as avgLatency, max(@duration) as maxLatency by bin(1h)