**Why It Matters (Architectural Logic)**: Executor sizing directly impacts job runtime and cost. Over-provision = waste; under-provision = OOM and retries. Balance parallelism vs. overhead.
Reserve ~1 executor per worker for OS/daemons. Per node: ~100GB RAM, 25 cores. Typical: 1 core = 1 executor. Executors per node: 4-5 (each ~4-5 cores, ~16-20GB). Leave 1 executor for driver/YARN. Example: 10 workers × 4 executors = 40 executors. Each executor: 4 cores, 16GB (4GB overhead)....
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 Meesho. 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.