# Cloning Databases with Hybrid Tables in Snowflake: A Deep Dive

> Cloning databases in Snowflake that contain hybrid tables offers powerful capabilities for data restoration and environment hydration. This article explores the nuances of cloning hybrid tables, including performance considerations and common pitfalls.

**Category:** snowflake  
**Published:** 2026-05-10T04:57:24.141043Z  
**Canonical:** https://allaboutspark.com/posts/cloning-databases-hybrid-tables-snowflake
**Tags:** snowflake, hybrid tables, database cloning, data management

---

## Introduction

Cloning databases in Snowflake is a powerful feature that allows you to create exact replicas of your data environment. This capability is particularly useful for scenarios such as point-in-time restores and environment hydration, where you need to replicate a production database for development or testing purposes. However, when it comes to hybrid tables, the cloning process introduces some unique challenges and considerations that you need to be aware of.

Hybrid tables in Snowflake are designed to optimize for both transactional and analytical workloads, providing low latency and high throughput through a row-based storage engine. They support row-level locking and enforce constraints like primary and foreign keys, making them ideal for applications that require high concurrency and integrity [7]. Cloning these tables, however, is not as straightforward as cloning standard tables due to their distinct characteristics and storage mechanisms [1].

## Understanding Hybrid Table Cloning

When you clone a database that contains hybrid tables, it's important to understand that the operation is a size-of-data operation rather than a metadata-only operation, as is the case with standard tables. This means that the data is physically copied into the row store, which can significantly impact both the time and cost of the operation, especially for large tables [1]. The cloning process for hybrid tables is akin to optimized direct bulk loading, which involves copying data rather than just pointers to data [1].

Moreover, hybrid table clones must be created at the database level. Attempting to clone at the schema or table level will result in an error unless you use the `IGNORE HYBRID TABLES` parameter to skip these tables [1][2]. This parameter is crucial when you need to clone a schema or database without including hybrid tables, allowing you to bypass errors and proceed with cloning other objects [3].

## Walking Through the Cloning Process

To clone a database containing hybrid tables, you use the `CREATE DATABASE ... CLONE` command. This command requires you to specify the source database and the destination database name. If you want to include hybrid tables in the clone, you must ensure that no Time Travel parameters like `AT BEFORE`, `OFFSET`, or `STATEMENT` are used, as these are not supported for hybrid tables [1][2]. Instead, you can specify an `AT TIMESTAMP` parameter to clone the database at a specific point in time.

Here's a basic example of how you might execute such a command:

```sql
CREATE DATABASE new_db CLONE source_db AT (TIMESTAMP => '2023-10-01T00:00:00.000Z');
```

In this command, `new_db` is the cloned database created from `source_db` as it existed at the specified timestamp. This operation will include all tables, including hybrid tables, from the source database as they were at that point in time [1].

## Common Mistakes and Pitfalls

One common mistake when cloning databases with hybrid tables is attempting to use unsupported Time Travel parameters. As mentioned, only the `AT TIMESTAMP` parameter is supported for hybrid tables, and using others will result in errors or the hybrid tables being silently skipped [2][3].

Another pitfall is underestimating the compute and storage costs associated with cloning hybrid tables. Since the operation involves physically copying data, it can be time-consuming and expensive, particularly for large datasets. It's crucial to plan for these costs and ensure that your Snowflake account has sufficient resources allocated [1].

Additionally, it's important to note that the history of the source tables is not retained in the cloned tables. This means that while you can use Time Travel to view the history of the source tables, the cloned tables will only have a history from the point of cloning onward [1].

## When to Use Hybrid Table Cloning

Cloning databases with hybrid tables is particularly beneficial when you need to restore a database to a previous state or when you need to replicate a production environment for testing purposes. The ability to create an exact replica of your data environment can significantly streamline development and testing workflows.

However, due to the costs and potential complexities involved, cloning hybrid tables may not be the best choice for every scenario. If your use case involves frequent cloning operations or if the tables are exceptionally large, you might need to consider alternative strategies, such as using standard tables or optimizing your data architecture to minimize the need for cloning.

In conclusion, while cloning databases with hybrid tables in Snowflake offers powerful capabilities, it requires careful consideration of the associated costs and limitations. By understanding these nuances, you can effectively leverage this feature to enhance your data management and operational workflows.

---

## Sources

1. [Clone databases that contain hybrid tables | Snowflake Documentation](https://docs.snowflake.com/en/user-guide/tables-hybrid-clone)
2. [CREATE … CLONE command: Cloning databases and schemas that contain hybrid tables | Snowflake Documentation](https://docs.snowflake.com/en/release-notes/bcr-bundles/2024_08/bcr-1792)
3. [CREATE <object> … CLONE | Snowflake Documentation](https://docs.snowflake.com/en/sql-reference/sql/create-clone)
4. [Best practices for hybrid tables | Snowflake Documentation](https://docs.snowflake.com/en/user-guide/tables-hybrid-best-practices)
5. [Limitations and unsupported features for hybrid tables | Snowflake Documentation](https://docs.snowflake.com/en/user-guide/tables-hybrid-limitations)
6. [Databases, Tables and Views - Overview | Snowflake Documentation](https://docs.snowflake.com/en/guides-overview-db)
7. [Hybrid tables | Snowflake Documentation](https://docs.snowflake.com/en/user-guide/tables-hybrid)
8. [Create hybrid tables | Snowflake Documentation](https://docs.snowflake.com/en/user-guide/tables-hybrid-create)