Reviewed by Aditya Kumar · Last reviewed 2026-03-24
Architectural use: Glue = serverless Spark; Catalog; connectors. Challenges at scale: (1) DPU sizing—OOM or slowness. (2) Job bookmarks—slow on many partitions. (3) Small files—poor performance; compact first. (4) Data skew—stragglers; salting. (5) Cost—DPU-hours....
Architectural use: Glue = serverless Spark; Catalog; connectors. Challenges at scale: (1) DPU sizing—OOM or slowness. (2) Job bookmarks—slow on many partitions. (3) Small files—poor performance; compact first. (4) Data skew—stragglers; salting. (5) Cost—DPU-hours. Best practice: Right-size; partition + predicates; compact; handle skew; incremental; monitor.
Red Flag: Default DPU for 100GB+ jobs. Pro-Move: 'We profiled; 64 DPU for 500GB; salt on skewed key—runtime 45 min vs 4h; we compact small files weekly.'
According to DataEngPrep.tech, this is one of the most frequently asked Cloud/Tools interview questions, reported at 1 company. DataEngPrep.tech maintains an editor-reviewed database of 1,863 data engineering interview questions across 7 categories.