Athena queries over a 10 TB S3 data lake are slow and expensive — Athena bills per TB scanned. What design change most reduces both cost and latency?
A.Larger Athena workgroup with more memory
Athena is serverless — you can't allocate memory.
B.Partition data by date/region in S3 + convert to a columnar format like Parquet
Correct. Partitioning + columnar format reduces bytes scanned.
C.Add more S3 buckets for the same data
More buckets without partitioning doesn't reduce scan volume.
D.Increase the query timeout
Timeouts don't reduce data scanned.
Why this answer
Partition pruning makes Athena scan only relevant partitions; columnar formats (Parquet/ORC) let it read only needed columns. Together they can cut scanned data 10-100x — the core Athena cost optimization.
Correct answer: B · Last reviewed 2026-10-26
Official AWS references
Practice this interactively
Answer without seeing the solution first in the Data Stores & Lakes practice session.
Practice Data Stores & Lakes