**Driver OOM**: (1) `collect()` on large DF. (2) Schema inference on huge file. (3) Large broadcast. Fix: Avoid collect; use limit or write+read. Provide schema. Reduce broadcast threshold.
**Executor OOM**: (1) Data skew—one partition huge. (2) Too few partitions; each too large. (3) Spill disabled or disk full. Fix: Salting; repartition; enable spill; increase partitions.
**Why It Happens**: Memory < data per partition....
The complete answer continues with detailed implementation patterns, architectural trade-offs, and production-grade considerations. It covers performance optimization strategies, common pitfalls to avoid, and real-world examples from companies like PWC. The answer also includes follow-up discussion points that interviewers commonly explore.
Continue Reading the Full Answer
Unlock the complete expert answer with code examples, trade-offs, and pro tips - plus 1,863+ more.
Or upgrade to Platform Pro - $39
Engineers who used these answers got offers at
AmazonDatabricksSnowflakeGoogleMeta
According to DataEngPrep.tech, this is one of the most frequently asked Spark/Big Data interview questions, reported at 1 company. DataEngPrep.tech maintains a curated database of 1,863+ real data engineering interview questions across 7 categories, verified by industry professionals.