An error occurred (ProvisionedThroughputExceededException)

Workarounds

1. 90% success Switch to on-demand capacity mode for unpredictable workloads

   aws dynamodb update-table --table-name MyTable --billing-mode PAY_PER_REQUEST
   
   On-demand mode auto-scales to handle any traffic level. No capacity planning needed, but costs ~6x more per request than provisioned at steady state.

   Sources: https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/HowItWorks.ReadWriteCapacityMode.html

2. 88% success Implement exponential backoff for throttled requests

   import boto3
   from botocore.config import Config
   
   config = Config(retries={'max_attempts': 10, 'mode': 'adaptive'})
   dynamodb = boto3.resource('dynamodb', config=config)
   
   Adaptive mode automatically adjusts retry delays based on throttling response headers.

   Sources: https://docs.aws.amazon.com/general/latest/gr/api-retries.html

3. 85% success Review partition key design — hot partitions cause throttling even with capacity

   Check CloudWatch ConsumedReadCapacityUnits/ConsumedWriteCapacityUnits per partition:
   aws cloudwatch get-metric-statistics --namespace AWS/DynamoDB --metric-name ConsumedWriteCapacityUnits --dimensions Name=TableName,Value=MyTable --period 300 --statistics Sum --start-time 2024-01-01T00:00:00Z --end-time 2024-01-01T01:00:00Z
   
   Use high-cardinality partition keys (e.g., userId, orderId) instead of low-cardinality keys (e.g., status, date). Add a random suffix to spread writes: partitionKey = userId + '#' + random(1,10).

   Sources: https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/bp-partition-key-design.html