Apache Iceberg is an open table format for very large analytic data sets. Iceberg supports modern analytical data operations such as record-level insert, update, delete, time-travel queries, ACID transactions, hidden partitioning and full schema evolution. It supports multiple underlying file storage formats such as Apache Parquet, Apache ORC and Apache Avro. Many data-processing engines support Apache Iceberg, including SQL engines such as Dremio and Trino as well as (structured) streaming engines such as Apache Spark and Apache Flink.
Apache Iceberg falls in the same category as Delta Lake and Apache Hudi. They all more or less support similar features, but each differs in the underlying implementations and detailed feature lists. Iceberg is an independent format and is not native to any specific processing engine, hence it's supported by an increasing number of platforms, including AWS Athena and Snowflake. For the same reason, Apache Iceberg, unlike native formats such as Delta Lake, may not benefit from optimizations when used with Spark.