BigData / Apache Iceberg Interview questions
What is the difference between Apache Iceberg and Apache Hudi?
Both are open table formats, but Hudi — originating at Uber — was built from the ground up around high-frequency upserts and streaming ingestion, while Iceberg's design historically prioritized simplicity and broad analytical query performance, leading to different strengths.
| Apache Iceberg | Apache Hudi |
| Snapshot-based metadata tree; simpler operating model. | Timeline-based architecture with file groups and record-level indexing. |
| Optimistic concurrency control only. | Supports OCC, MVCC, and non-blocking concurrency control. |
| Strong general analytical scan performance. | 10-30x faster point lookups via record-level indexing; leads for streaming upserts. |
| Broadest multi-engine and catalog ecosystem support. | Deep native integration with streaming tools like Flink and Kafka Streams. |
The general guidance that's emerged is workload-dependent: Iceberg tends to be favored for building a broad, vendor-neutral analytical lakehouse with strong multi-engine support, while Hudi tends to be favored specifically for high-frequency, mutable, streaming-heavy workloads where its record-level indexing and purpose-built table services (compaction, clustering, cleaning) provide a meaningful edge.
What architecture does Hudi use, distinct from Iceberg's snapshot tree?
What kind of workload does Hudi tend to be favored for?
More Related questions...
