Glossary, read 10 Nov 2026
Spark acceleration terms, each with a sourced example
Open the LedgerCompare two options
35 terms, A to Z
- Adaptive Query Execution
- Apache Arrow
- APU (Analytics Processing Unit)
- AWS F2 instance
- Broadcast hash join
- Catalyst
- Columnar batch
- Cost per job
- DataFusion
- Dynamic partition pruning
- EMR on EKS
- EMR release label
- Executor
- FPGA
- GPU
- Iceberg compaction (rewrite_data_files)
- Off-heap memory
- Operator fallback
- Parquet row group
- Partition
- Scale factor
- Shuffle
- Skew
- Sort-merge join
- Spark Operator
- Spark plugin (spark.plugins)
- Speculative execution
- Spill
- Substrait
- TPC-DS
- TPC-H
- Tungsten
- Vectorized execution
- Velox
- Whole-stage code generation
Last reviewed 10 Nov 2026.