# Lightning Tables Support and Limitations in Confluent Cloud Lightning Tables support Basic, Standard, Enterprise, and Dedicated clusters on AWS, for topics with Avro, Protobuf, or JSON Schema data. For supported data types and SQL statements, see [SQL Reference for Lightning Table Queries in Confluent Cloud](query/sql-reference.md#lightning-tables-sql-reference). ## Supported environments Lightning Tables are available on the following cluster types on AWS: - Basic - Standard - Enterprise - Dedicated For the list of supported regions, see [Cloud Regions](../get-started/regions.md#providers-regions). ## Data model requirements To enable a Lightning Table for a topic, the topic must have a registered schema in Avro, Protobuf, or JSON Schema format. ## Upsert mode A Lightning Table automatically selects the ingestion mode based on the topic’s [cleanup.policy](../topics/manage.md#topics-cleanup-policy) configuration. In upsert mode, the Lightning Table deduplicates records by key so that it reflects the latest state for each key. This behavior is consistent with Apache Kafka® log compaction. ### Upsert requirements To use upsert mode, your topic must meet the following requirements in addition to the [general data model requirements](#rtce-data-model-requirements): - The topic’s `cleanup.policy` must include `compact`. Topics with the default `delete` policy use append mode. - The topic must have a Kafka record key, which the Lightning Table uses as the primary key for deduplication. The key can be raw bytes, or it can use a schema in Avro, Protobuf, or JSON Schema format. A Lightning Table doesn’t support changes to the key schema. For more information, see [Schema Evolution for Lightning Tables in Confluent Cloud](schema-evolution.md#lightning-tables-schema-evolution). ### Upsert behavior Upsert mode has the following behavior: - **Last-write-wins**: When multiple records share the same key, the Lightning Table keeps the most recent record and replaces earlier records with the same key. - **Deletes**: A record with a null value, known as a tombstone, deletes its key from the Lightning Table. A later record with the same key adds it back. ## Related content - [Lightning Tables in Confluent Cloud](overview.md#lightning-tables-overview) - [SQL Reference for Lightning Table Queries in Confluent Cloud](query/sql-reference.md#lightning-tables-sql-reference) - [Schema Evolution for Lightning Tables in Confluent Cloud](schema-evolution.md#lightning-tables-schema-evolution) - [Private Networking for Lightning Tables in Confluent Cloud](private-networking.md#rtce-private-networking) #### NOTE This website includes content developed at the [Apache Software Foundation](https://www.apache.org/) under the terms of the [Apache License v2](https://www.apache.org/licenses/LICENSE-2.0.html).