Amazon SimpleDB is not storing raw data. It transforms raw data into multiple indices and is then considered to be a column store database.
Pro/Cons
SimpleDB requires customers to manage their datasets in containers called Domains, which have a finite capacity in terms of storage (10 GB) and request throughput.
Indexes all attributes for each item stored in a domain:
pro: Simplifies the schema design and provides query flexibility
cons: No predictability particularly as dataset sizes grow.
No write predictability: every database write needs to update all attribute indices (regardless of whether the customer is using all the indices for querying).
No read predictability: working set does not always fit in memory.