TRUNCATE TABLE statement was used to remove all data, the IDENTITY column resets the numbering to start with the original value. DELETE statement was used to remove all data, the IDENTITY values are still incremented according to the last value used.

A TRUNCATE TABLE statement tends to use fewer locks than a DELETE statement. When a TRUNCATE TABLE statement is used, SQL Server applies table and page locks but not row locks, as a DELETE statement does.

A TRUNCATE TABLE statement uses less transaction log space than a DELETE statement. When a TRUNCATE TABLE statement is used, SQL Server de-allocates the data pages and records only the de-allocations in the transaction log. When a DELETE statement is used, SQL Server makes an entry into the transaction log for each deleted row.

A TRUNCATE TABLE statement leaves no pages in a table, whereas a DELETE statement can leave empty pages.

While a TRUNCATE TABLE statement is more efficient than a DELETE statement, there are restrictions that govern the use of TRUNCATE TABLE. For example, you should not use a TRUNCATE TABLE statement against a table under the following conditions:

An indexed view specifies the table in its definition.

Transaction or merge replication is used to publish the table.

The table is referenced by a foreign key constraint (unless it is a self-referencing foreign key).

The table's IDENTITY values must be preserved and consistently incremented.

Only specific rows are to be deleted from the table, and not the entire dataset.

Regardless of these differences, both the TRUNCATE TABLE and DELETE statements remove only data and do not impact the table structure. Indexes, constraints, and columns are left untouched.